Patient, clinical, surgical, and institutional factors associated with length of stay in scheduled degenerative thoracolumbar spine surgery: National Multicenter Cohort Analysis from the Canadian Spine Outcomes and Research Network

Charlotte Dandurand Combined Neurosurgical and Orthopaedic Spine Program, University of British Columbia, Vancouver, British Columbia;

Search for other papers by Charlotte Dandurand in
jns
Google Scholar
PubMed
Close
 MD, MSc
,
Mohammad S. Mashayekhi Combined Neurosurgical and Orthopaedic Spine Program, University of British Columbia, Vancouver, British Columbia;

Search for other papers by Mohammad S. Mashayekhi in
jns
Google Scholar
PubMed
Close
 PhD
,
Greg McIntosh Canadian Spine Outcomes and Research Network, Canadian Spine Society, Markdale, Ontario;

Search for other papers by Greg McIntosh in
jns
Google Scholar
PubMed
Close
 MSc
,
John T. Street Combined Neurosurgical and Orthopaedic Spine Program, University of British Columbia, Vancouver, British Columbia;

Search for other papers by John T. Street in
jns
Google Scholar
PubMed
Close
 MD
,
Charles G. Fisher Combined Neurosurgical and Orthopaedic Spine Program, University of British Columbia, Vancouver, British Columbia;

Search for other papers by Charles G. Fisher in
jns
Google Scholar
PubMed
Close
 MD, MHSc
,
Joel Finkelstein Department of Surgery, Sunnybrook Health Sciences Centre, Toronto, Ontario;

Search for other papers by Joel Finkelstein in
jns
Google Scholar
PubMed
Close
 MD
,
Edward Abraham Department of Surgery, Dalhousie University, Halifax, Nova Scotia;

Search for other papers by Edward Abraham in
jns
Google Scholar
PubMed
Close
 MD
,
Jérôme Paquet Centre Hospitalier Universitaire de Québec, Hôpital Enfant-Jésus, Québec City, Québec;

Search for other papers by Jérôme Paquet in
jns
Google Scholar
PubMed
Close
 MD
,
Hamilton Hall Department of Surgery, University of Toronto, Ontario;

Search for other papers by Hamilton Hall in
jns
Google Scholar
PubMed
Close
 MD
,
Eugene Wai Department of Surgery, Ottawa Hospital, Ottawa, Ontario;

Search for other papers by Eugene Wai in
jns
Google Scholar
PubMed
Close
 MD
,
Daryl R. Fourney Department of Surgery, University of Saskatchewan, Saskatoon, Saskatchewan;

Search for other papers by Daryl R. Fourney in
jns
Google Scholar
PubMed
Close
 MD
,
Christopher S. Bailey London Health Sciences Centre, Combined Neurosurgical and Orthopaedic Spine Program, Western University, London, Ontario;

Search for other papers by Christopher S. Bailey in
jns
Google Scholar
PubMed
Close
 MD
,
Sean D. Christie Department of Surgery, Dalhousie University, Halifax, Nova Scotia;

Search for other papers by Sean D. Christie in
jns
Google Scholar
PubMed
Close
 MD
,
Alex Soroceanu Department of Surgery, University of Calgary, Alberta;

Search for other papers by Alex Soroceanu in
jns
Google Scholar
PubMed
Close
 MD
,
Michael Johnson Winnipeg Health Sciences Centre, University of Manitoba, Winnipeg, Manitoba;

Search for other papers by Michael Johnson in
jns
Google Scholar
PubMed
Close
 MD
,
Adrienne Kelly Department of Surgery, Northern Ontario School of Medicine, Sault Ste. Marie, Ontario;

Search for other papers by Adrienne Kelly in
jns
Google Scholar
PubMed
Close
 MD
,
Travis E. Marion Department of Surgery, Thunder Bay Regional Health Science Centre, Thunder Bay, Ontario;

Search for other papers by Travis E. Marion in
jns
Google Scholar
PubMed
Close
 MD
,
Andrew Nataraj Department of Surgery, University of Alberta Hospital, Edmonton, Alberta;

Search for other papers by Andrew Nataraj in
jns
Google Scholar
PubMed
Close
 MD
,
Carlo Santaguida Department of Surgery, McGill University, Montréal, Québec;

Search for other papers by Carlo Santaguida in
jns
Google Scholar
PubMed
Close
 MD
,
Daniel Warren Department of Neurosurgery, Vancouver Island Health Authority, Victoria, British Columbia; and

Search for other papers by Daniel Warren in
jns
Google Scholar
PubMed
Close
 MD
,
Thomas Guy Hogan Department of Orthopaedic Surgery, Health Sciences Centre, St. John’s, Newfoundland and Labrador, Canada

Search for other papers by Thomas Guy Hogan in
jns
Google Scholar
PubMed
Close
 MD
,
Neil Manson Department of Surgery, Canada East Spine Centre, Saint John, New Brunswick;

Search for other papers by Neil Manson in
jns
Google Scholar
PubMed
Close
 MD
,
Philippe Phan Department of Surgery, Ottawa Hospital, Ottawa, Ontario;

Search for other papers by Philippe Phan in
jns
Google Scholar
PubMed
Close
 MD, PhD
,
Henry Ahn Department of Surgery, University of Toronto, Ontario;

Search for other papers by Henry Ahn in
jns
Google Scholar
PubMed
Close
 MD
,
Y. Raja Rampersaud University Health Network, Toronto Western Hospital, Toronto, Ontario;

Search for other papers by Y. Raja Rampersaud in
jns
Google Scholar
PubMed
Close
 MD
,
Jocelyn Blanchard Department of Surgery, Centre Hospitalier Universitaire de Sherbrooke, Sherbrooke, Québec;

Search for other papers by Jocelyn Blanchard in
jns
Google Scholar
PubMed
Close
 MD
,
Kenneth Thomas Department of Surgery, University of Calgary, Alberta;

Search for other papers by Kenneth Thomas in
jns
Google Scholar
PubMed
Close
 MD
,
Nicolas Dea Combined Neurosurgical and Orthopaedic Spine Program, University of British Columbia, Vancouver, British Columbia;

Search for other papers by Nicolas Dea in
jns
Google Scholar
PubMed
Close
 MD, MSc
, and
Raphaële Charest-Morin Combined Neurosurgical and Orthopaedic Spine Program, University of British Columbia, Vancouver, British Columbia;

Search for other papers by Raphaële Charest-Morin in
jns
Google Scholar
PubMed
Close
 MD
Free access

OBJECTIVE

Length of stay (LOS) is a contributor to costs and resource utilization. The primary goal of this study was to identify patient, clinical, surgical, and institutional variables that influence LOS after elective surgery for thoracolumbar degenerative pathology. The secondary objective was to examine variability in LOS and institutional strategies used to decrease LOS.

METHODS

This is a retrospective study of prospectively collected data from a multicentric cohort enrolled in the Canadian Spine Outcomes and Research Network (CSORN) between January 2015 and October 2020 who underwent elective thoracolumbar surgery (discectomy [1 or 2 levels], laminectomy [1 or 2 levels], and posterior instrumented fusion [up to 5 levels]). Prolonged LOS was defined as LOS greater than the median. Logistic regression models were used to determine factors associated with prolonged LOS for each procedure. A survey was sent to the principal investigators of the participating healthcare institutions to understand institutional practices that are used to decrease LOS.

RESULTS

A total of 3700 patients were included (967 discectomies, 1094 laminectomies, and 1639 fusions). The median LOSs for discectomy, laminectomy, and fusion were 0.0 (IQR 1.0), 1.0 (IQR 2.0), and 4.0 (IQR 2.0) days, respectively. On multivariable analysis, predictors of prolonged LOS for discectomy were having more leg pain, higher Oswestry Disability Index (ODI) scores, symptom duration more than 2 years, having undergone an open procedure, occurrence of an adverse event (AE), and treatment at an institution without protocols to reduce LOS (p < 0.05). Predictors of prolonged LOS for laminectomy were increased age, living alone, higher ODI scores, higher BMI, open procedures, longer operative time, AEs, and treatment at an institution without protocols to reduce LOS (p < 0.05). For posterior instrumented fusion, predictors of prolonged LOS were older age, living alone, more comorbidities, higher ODI scores, longer operative time, AEs, and treatment at an institution without protocols to reduce LOS (p < 0.05). The laminectomy group had the largest variability in LOS (SD 4.4 days, range 0–133 days). Three hundred fifty-four patients (22%) had an LOS above the 75th percentile. Ten institutions (53%) had either Enhanced Recovery After Surgery or standardized protocols in place.

CONCLUSIONS

Among the factors identified in this study, worse baseline ODI scores, experiencing AEs, and treatment at an institution without protocols aimed at reducing LOS were predictive of prolonged LOS in all surgical groups. The laminectomy group had the largest variability in LOS.

ABBREVIATIONS

AE = adverse event; AUC = area under the ROC curve; CSORN = Canadian Spine Outcomes and Research Network; ERAS = Enhanced Recovery After Surgery; LOS = length of stay; MCS = Mental Component Summary; MIS = minimally invasive surgery; NRS = numeric rating scale; ODI = Oswestry Disability Index; PCS = Physical Component Summary; PHQ-9 = Patient Health Questionnaire–9; ROC = receiver operating characteristic; SAVES = Spine Adverse Events Severity System; SF-12 = 12-item Short-Form Health Survey.

OBJECTIVE

Length of stay (LOS) is a contributor to costs and resource utilization. The primary goal of this study was to identify patient, clinical, surgical, and institutional variables that influence LOS after elective surgery for thoracolumbar degenerative pathology. The secondary objective was to examine variability in LOS and institutional strategies used to decrease LOS.

METHODS

This is a retrospective study of prospectively collected data from a multicentric cohort enrolled in the Canadian Spine Outcomes and Research Network (CSORN) between January 2015 and October 2020 who underwent elective thoracolumbar surgery (discectomy [1 or 2 levels], laminectomy [1 or 2 levels], and posterior instrumented fusion [up to 5 levels]). Prolonged LOS was defined as LOS greater than the median. Logistic regression models were used to determine factors associated with prolonged LOS for each procedure. A survey was sent to the principal investigators of the participating healthcare institutions to understand institutional practices that are used to decrease LOS.

RESULTS

A total of 3700 patients were included (967 discectomies, 1094 laminectomies, and 1639 fusions). The median LOSs for discectomy, laminectomy, and fusion were 0.0 (IQR 1.0), 1.0 (IQR 2.0), and 4.0 (IQR 2.0) days, respectively. On multivariable analysis, predictors of prolonged LOS for discectomy were having more leg pain, higher Oswestry Disability Index (ODI) scores, symptom duration more than 2 years, having undergone an open procedure, occurrence of an adverse event (AE), and treatment at an institution without protocols to reduce LOS (p < 0.05). Predictors of prolonged LOS for laminectomy were increased age, living alone, higher ODI scores, higher BMI, open procedures, longer operative time, AEs, and treatment at an institution without protocols to reduce LOS (p < 0.05). For posterior instrumented fusion, predictors of prolonged LOS were older age, living alone, more comorbidities, higher ODI scores, longer operative time, AEs, and treatment at an institution without protocols to reduce LOS (p < 0.05). The laminectomy group had the largest variability in LOS (SD 4.4 days, range 0–133 days). Three hundred fifty-four patients (22%) had an LOS above the 75th percentile. Ten institutions (53%) had either Enhanced Recovery After Surgery or standardized protocols in place.

CONCLUSIONS

Among the factors identified in this study, worse baseline ODI scores, experiencing AEs, and treatment at an institution without protocols aimed at reducing LOS were predictive of prolonged LOS in all surgical groups. The laminectomy group had the largest variability in LOS.

In Brief

Efforts have been directed at decreasing length of stay (LOS) to reduce healthcare cost. Researchers identified patient, clinical, surgical, and institutional factors predictive of prolonged LOS after three common elective degenerative thoracolumbar spine surgeries. In all groups, worse baseline Oswestry Disability Index scores, experiencing adverse events, and being treated at an institution without protocols to reduce LOS were predictive of prolonged LOS. This study suggests that institutional protocols such as an Enhanced Recovery After Surgery pathway, outpatient surgery, and routine minimally invasive surgery may be effective at decreasing LOS, but further investigation is needed to determine which measures are beneficial.

Degenerative spinal pathologies are common and carry a significant economic burden.1 Over recent decades, the number of spinal surgeries performed has increased with a significant increase in resource usage, in part because of the rapidly aging population.24 In 2018, The Lancet published a series on low-back pain and demanded a call for action in curtailing healthcare costs.5 Length of stay (LOS) is an important outcome measure for stakeholders who advocate for optimal use of limited healthcare resources.6

In recent years, efforts have been directed at decreasing LOS as a means of reducing healthcare costs. Various Enhanced Recovery After Surgery (ERAS) protocols have been implemented and have been associated with a reduced incidence of complications and LOS.79 A thorough analysis of the association between institutional measures included in ERAS protocols and LOS after lumbar spine surgery has not been performed in Canada. In Canada, while having a universal healthcare system, healthcare policies are not under federal legislation, and as such, the implementation of ERAS pathways has been variable among its healthcare institutions. Additionally, it is perceived by the spinal community that LOS varies tremendously across healthcare institutions in Canada. How this variability differs among common lumbar surgical procedures is unknown, and factors explaining this variability have not been evaluated but are crucial to guide future efforts to optimize limited healthcare resources and associated costs.

The primary objective of this study was to identify patient, clinical, surgical, and institutional factors predictive of prolonged LOS after three common elective degenerative thoracolumbar spine surgeries: discectomy, laminectomy, and posterior instrumented fusion. The secondary objective was to examine variability in LOS and institutional practices aimed at reducing LOS among healthcare institutions that enrolled patients in the current study.

Methods

Study Design

This is a retrospective analysis of prospectively collected data from the multicenter Canadian Spine Outcomes and Research Network (CSORN) registry. CSORN is a national registry that includes 75 neurosurgical and orthopedic spine surgeons from 19 institutions (https://www.csorncss.ca/).10 Since 2012, more than 16,600 patients have been enrolled. Standardized data collection is performed in the preoperative and postoperative periods at prespecified time points. The CSORN registry does not capture institutional practice patterns. A survey was sent to the principal investigator at each of the 19 participating institutions. Local research ethics board approvals were obtained prior to data collection.

Patient Sample

This study includes all consecutive patients enrolled in the CSORN registry between January 2015 and October 2020. Inclusion criteria for this study were patients 18 years of age or older who underwent elective thoracolumbar spine surgery for degenerative conditions with reported LOS data. Surgeries included were discectomy (1 or 2 levels), laminectomy (1 or 2 levels), and posterior instrumentation fusion (up to 5 levels, with or without decompression, with or without interbody fusion). Exclusion criteria were missing LOS data, revision surgery, emergency surgery, deformity diagnosis, cervical location, and combined anterior-posterior approach for the posterior instrumented fusion group.

Patient, Clinical, and Surgical Variables Considered

The independent variables used for predictive modeling were measured using either categorical or continuous scales as appropriate. Baseline preoperative patient characteristics included sociodemographic factors (sex, age, living status, BMI, number of comorbidities, tobacco use, and education level). Clinical variables were symptom duration, medication usage (pain, narcotics, and antidepressants), and preoperative patient-reported outcome measures (numeric rating scale [NRS] back and leg,11 Oswestry Disability Index [ODI],12 12-item Short-Form Health Survey [SF-12] Mental Component Summary [MCS], SF-12 Physical Component Summary [PCS],13 and Patient Health Questionnaire–9 [PHQ-9]14 scores). Surgical variables included whether the procedure was minimally invasive surgery (MIS) or open, operative time, and the occurrence of any intra- and postoperative adverse events (AEs). AEs are prospectively collected using the Spine Adverse Events Severity System (SAVES) version 2, containing 14 specific intraoperative and 22 postoperative AEs.15

Institutional Variables Considered

A survey was sent to the principal investigator at each of the 19 participating healthcare institutions to determine measures in 1) preoperative spine education sessions, 2) identification of potential discharge obstacles, 3) written information about expected LOS and discharge instructions, and 4) preoperative contract. The survey included the following routine healthcare practices: Foley catheter insertion and removal, usage of patient-controlled anesthesia, routine postoperative assessment by physiotherapists and occupational therapists, and the presence of a combined orthopedic and neurosurgical unit. Institutions were asked if they had an ERAS in place (yes/no). Institutions that answered "yes" were asked to provide details about the initiatives taken to reduce LOS. Open questions were asked to explore what future efforts should be put in place to target perceived barriers to timely discharge. Institutions were categorized as high enrollers (> 100 patients) and low enrollers (≤ 100 patients) to allow institutional volume comparison between institutions that have an ERAS or standardized protocols in place and institutions that do not.

Outcome Measures

LOS was defined as the number of calendar days from the operation date to the discharge date from the hospital where the surgery took place. An overnight stay was considered as a 1-day LOS. Given the right-skewed (positive) distribution of the LOS data, the outcome chosen was the binary classification of whether LOS was at or below versus above the median for each procedure (discectomy, laminectomy, and posterior instrumented fusion). The median was chosen over the mean to mitigate the outlier’s effect.

Statistical Analysis

The Transparent Reporting of a Multivariable Prediction Model for Individual Prognosis or Diagnosis (TRIPOD) guideline was followed for the development and validation of the prediction models.16

Baseline differences between patients with an LOS less than or equal to versus above the median LOS were compared using Student t-tests for continuous variables and chi-square tests for categorical factors. The survey results were reported with descriptive statistics.

Prolonged LOS (LOS greater than the median) was defined for the discectomy, laminectomy, and posterior instrumented fusion groups. Logistic regression models were used to determine patient, clinical, surgical, and institutional factors associated with prolonged LOS for each procedure. Potential predictors were grouped into the following three clusters: 1) patient characteristic variables (sex, age, living arrangement, BMI, number of comorbidities, tobacco use, and education level); 2) clinical factors (symptom duration; medication usage [pain, narcotics, and antidepressants]; and NRS back and leg pain, ODI, SF-12 MCS, SF-12 PCS, and PHQ-9 scores); and 3) surgical and institutional variables (MIS, operative time, AEs, and protocol in place to reduce LOS). To assess variation in institutional practices, the dichotomous variable of having protocols in place aimed at reducing LOS (regardless of the number of measures put in place) was included in the model. Multivariable logistic regression was used to develop a predictive model to predict the relationship between the outcome of prolonged LOS and the independent variables of interest. A backward conditional selection procedure was used for each cluster. Significant variables from each cluster modeling procedure were combined to create the best multivariate model predicting outcome. The presence of collinearity was assessed, and estimated blood loss and the number of levels were not included in the multivariable analysis because of collinearity with operative time duration. Bootstrap samples were repeatedly drawn from the data set. There were 1000 bootstrap samples created to internally validate the final models. An alpha of p < 0.10 was used for model entry and p < 0.15 for exit, given that for decisions on entering and omitting terms, the significance level should not be too small or too rigid.17 A receiver operating characteristic (ROC) curve was created for each group to assess discriminatory ability to predict prolonged LOS. An area under the ROC curve (AUC) of 0.5 suggested no discrimination, 0.7 to 0.8 was considered acceptable, 0.8 to 0.9 was considered excellent, and more than 0.9 was considered outstanding.18,19

All analyses were conducted using IBM SPSS Statistics for Windows (version 28, IBM Corp.).

Results

A total of 3700 patients met the inclusion criteria. Of the patients included in the analysis, 967 patients underwent discectomy, 1094 patients underwent laminectomy, and 1639 patients underwent posterior instrumented fusion (Fig. 1). Patients with missing LOS data were excluded from analysis (n = 37 for discectomy, n = 52 for laminectomy, and n = 84 for posterior instrumented fusion).

FIG. 1.
FIG. 1.

Flowchart of patient inclusion and exclusion. TL = thoracolumbar. *Patients were analyzed in the laminectomy group.

Patients included in this study were treated at 19 institutions by 75 spine surgeons. All surgeons performed discectomy, 73 performed laminectomy, and 66 performed posterior instrumented fusion. The mean number of enrolling surgeons per institution was 3.9 (range 1–11). The mean number of enrolled patients per institution was 194.7 (range 5–1123), and the mean number of enrolled patients per surgeon was 49.3 (range 1–258). For the discectomy group, the mean number of enrolled patients per institution was 60.4 (range 2–276), and the mean number of enrolled patients per surgeon was 15.9 (range 1–120). For the laminectomy group, the mean number of enrolled patients per institution was 57.6 (range 2–411), and the mean number of enrolled patients per surgeon was 17.6 (range 1–101). For the posterior fusion group, the mean number of enrolled patients per institution was 91.1 (range 2–436), and the mean number of enrolled patients per surgeon was 24.8 (range 1–128). All institutions enrolled patients undergoing the three procedures. High-enrollment institutions (> 100 patients) were more likely to have protocols in place to reduce LOS compared with low-enrollment institutions (≤ 100 patients) (74% vs 33%, p < 0.001).

For the discectomy group, the most common level operated on was L5–S1 (50.2%), followed by L4–5 (36.7%) and L3–4 (3.4%). For the laminectomy group, the most common level operated on was L4–5 (32.2%), followed by L3-4-5 (15.4%) and L3–4 (11.4%). For the posterior instrumented fusion group, the most common level operated on was L4–5 (43.1%), followed by L–S1 (16.9%) and L3-4-5 (7.2%). In all three groups, the most common AE was dural tear, followed by urinary retention. The most common intra- and postoperative AEs are detailed in Appendix 3.

Patients, Pathology, and Surgical Variables

Discectomy Group

The median LOS was 0 (IQR 1) days. Of the 967 patients, 602 (62.2%) had the surgery in an outpatient setting. The discectomy group had the smallest variability in LOS (SD 1.9 days, range 0–41 days). The frequency distribution of the median LOS is shown in Fig. 2A. Sixty-nine patients (7%) had an LOS above the 75th percentile.

FIG. 2.
FIG. 2.

Frequency distribution of the median LOS per procedure. A: Discectomy. B: Laminectomy. C: Posterior thoracolumbar fusion.

Table 1 displays the sample characteristics, overall and by median. On univariate analysis, patients who experienced prolonged LOS were older; used tobacco products less often; had longer symptom duration; more commonly had a chief complaint of back pain or neurogenic claudication but less commonly radiculopathy; and had worse baseline NRS back pain, ODI, and SF-12 PCS scores (p < 0.05). Patients with prolonged LOS underwent fewer MIS procedures, had a longer operative duration and more AEs, and were more likely to be treated at an institution without protocols in place to reduce LOS (p < 0.05).

TABLE 1.

Discectomy group: univariate analysis

Overall (n = 967)LOS 0 Days (n = 598)LOS ≥1 Day (n=369)p Value
Patient/clinical variables
 Male sex, %53.654.851.50.309
 Mean (SD) age, yrs44.3 (13.3)43.5 (12.8)45.5 (14.1) 0.020
 Living arrangement, % (n = 953)

   Living alone

   Cohabitation


 12.8

 87.2


 13.1

 86.9


 12.3

 87.7
0.196
 Mean (SD) BMI (n = 947)27.8 (5.3)27.7 (5.3)27.9 (5.4)0.659
 Mean (SD) no. of comorbidities (n = 941)2.0 (1.6)2.0 (1.6)2.1 (1.5)0.164
 Tobacco use, % (n = 948)22.625.018.7 0.024
 Education ≤ high school, % (n = 947)30.630.431.00.842
 Symptom duration >2 yrs, % (n = 963)34.829.742.9 <0.001
 Pain medication use, % (n = 961)91.191.690.20.463
 Narcotic use, % (n = 854)64.163.065.80.397
 Antidepressant use, % (n = 839)23.621.826.50.123
 Chief complaint, n (%) (n = 965)

   Back pain

   Neurogenic claudication

   Radiculopathy

   Other


 104 (10.8)

 14 (1.4)

 842 (87.1)

 5 (0.5)


 39 (6.5)

 3 (0.5)

 552 (92.3)

 2 (0.3)


 65 (17.6)

 11 (3.0)

 290 (78.6)

 3 (0.8)


<0.001

0.003

<0.001

0.376
 Mean (SD) preop PROs

   NRS back pain (n = 961)

   NRS leg pain (n = 962)

   ODI (n = 961)

   SF-12 MCS (n = 942)

   SF-12 PCS (n = 942)

   PHQ-9 (n = 966)


 6.46 (2.47)

 7.33 (2.19)

 49.0 (16.2)

 41.2 (11.2)

 30.0 (7.7)

 10.9 (6.0)


 6.31 (2.52)

 7.43 (1.95)

 47.6 (15.7)

 41.3 (10.7)

 30.6 (7.6)

 10.7 (5.8)


 6.70 (2.38)

 7.16 (2.52)

 51.4 (16.9)

 41.0 (12.0)

 29.0 (7.8)

 11.2 (6.2)


0.016

0.083

<0.001

0.702

0.003

0.214
Surgical/institutional variables
 MIS, %61.576.137.9 <0.001
 Mean (SD) OR duration, mins (n = 905)88.5 (96.0)78.9 (101.4)104.1 (84.3) <0.001
 Intra-/postop AEs, %7.72.716.3 <0.001
 Protocol to reduce LOS, % (n = 887)72.876.866.3 <0.001
 Mean (SD) no. of operated levels1.06 (0.24)1.06 (0.23)1.07 (0.25)0.212

OR = operating room; PRO = patient-reported outcome.

Percentages are based on the total number of patients in each column. Boldface type indicates statistical significance.

On multivariable analysis, there were 6 predictors of prolonged LOS (Table 2). Patients treated at institutions without protocols in place to reduce LOS and those with a symptom duration longer than 2 years had almost twice the odds of prolonged LOS; open (non-MIS) procedures and having AEs had more than 5 times the odds of prolonged LOS. For each 1-point decrease in the NRS leg pain score, the odds of prolonged LOS decreased by almost 10%. For each 10-point increase in ODI, the odds of prolonged LOS increased by 30%. The AUC was 0.78 (95% CI 0.75–0.81, p < 0.001), indicating acceptable discriminatory ability of this predictive model (Appendix 1).

TABLE 2.

Discectomy group: multivariable logistic regression model for predictive factors of prolonged LOS (> 0 days)

OR95% CIp Value
Patient/clinical variables
 Decreased NRS leg pain0.9050.833–0.982  0.017
 Higher ODI1.031.02–1.04 <0.001
 Symptom duration >2 yrs1.771.26–2.48 <0.001
Surgical/institutional variables
 Open procedures6.224..49–8.62 <0.001
 Having intra-/postop AEs7.854.09–15.06 <0.001
 No protocol to reduce LOS1.931.36–2.74 <0.001

Boldface type indicates statistical significance.

Laminectomy Group

The median LOS in the laminectomy group was 1 (IQR 2) day. Of the 1094 patients, 680 (62%) had an LOS of 1 day or less. Of the three procedure groups, the laminectomy group had the largest variability in LOS (SD 4.4 days, range 0–133 days). The frequency distribution of the median LOS is shown in Fig. 2B. Fifty-four patients (5%) had an LOS above the 75th percentile.

Table 3 displays the sample characteristics, overall and by median. On univariate analysis, patients who experienced prolonged LOS were older, more commonly lived alone, had more comorbidities, more commonly had a chief complaint of neurogenic claudication and less commonly radiculopathy, and had worse baseline ODI and SF-12 MCS and PCS scores (p < 0.05). Patients with prolonged LOS underwent MIS procedures less frequently, had longer operative durations and more AEs, and were more likely to be treated at an institution without protocols in place to reduce LOS (p < 0.05).

TABLE 3.

Laminectomy group: univariate analysis

Overall (n = 1094)LOS ≤1 Day (n = 680)LOS >1 Day (n = 414)p Value
Patient/clinical factors
 Male sex, %58.659.457.2 0.481
 Mean (SD) age, yrs66.1 (11.8)64.1 (11.9)69.4 (10.9) <0.001
 Living arrangement, % (n = 1089)

   Living alone

   Cohabitation


 17.3

 82.7


 13.4

 86.6


 23.8

 76.2
<0.001
 Mean (SD) BMI (n = 1078)28.9 (5.3)28.8 (5.1)29.2 (5.6)0.181
 Mean (SD) no. of comorbidities (n = 1078)3.0 (1.8)2.9 (1.8)3.2 (1.8) 0.038
 Tobacco use, % (n = 1084)12.413.710.20.094
 Education ≤ high school, % (n = 1072)42.040.943.80.354
 Symptom duration >2 yrs, % (n = 1091)64.465.462.80.378
 Pain medication use, % (n = 1093)87.386.987.90.637
 Narcotic use, % (n = 922)47.346.049.30.337
 Antidepressant use, % (n = 901)19.919.520.50.698
 Chief complaint, n (%) (n = 1080)

   Back pain

   Neurogenic claudication

   Radiculopathy

   Other


 41 (3.7)

 599 (54.8)

 436 (39.9)

 4 (0.4)


 22 (3.2)

 344 (50.6)

 306 (45.0)

 2 (0.3)


 19 (4.6)

 255 (61.6)

 130 (31.4)

 2 (0.5)


 0.253

<0.001

<0.001

0.636
 Mean (SD) preop PROs

   NRS back pain (n = 1090)

   NRS leg pain (n = 1087)

   ODI

   SF-12 MCS (n = 1043)

   SF-12 PCS (n = 1043)

   PHQ-9


 6.60 (2.48)

 7.27 (2.14)

 44.3 (15.5)

 45.4 (11.4)

 29.3 (7.8)

 8.2 (5.9)


 6.56 (2.50)

 7.19 (2.15)

 43.0 (15.5)

 45.9 (11.3)

 30.0 (8.0)

 8.1 (6.0)


 6.66 (2.44)

 7.40 (2.11)

 46.5 (15.5)

 44.4 (11.5)

 28.2 (7.3)

 8.2 (5.7)


 0.512

 0.107

<0.001

0.034

<0.001

  of this predictive model 0.716
Surgical/institutional variables
 MIS, %49.158.833.1 <0.001
 Mean (SD) OR duration, mins (n = 1023)96.9 (50.0)85.1 (36.5)116.3 (61.9) <0.001
 Intra-/postop AEs, %18.37.935.3 <0.001
 Protocol to reduce LOS, % (n = 1058)65.869.959.0 <0.001
 Mean (SD) no. of operated levels1.26 (0.44)1.21 (0.41)1.33 (0.47) <0.001

Percentages are based on the total number of patients in each column. Boldface type indicates statistical significance.

On multivariable analysis, there were 8 predictors of prolonged LOS (Table 4). Patients living alone had approximately twice the odds of prolonged LOS. Those undergoing open (non-MIS) surgery and those treated at institutions without protocols to reduce LOS had approximately 3 times the odds of prolonged LOS. Patients with AEs had more than 5 times the odds of prolonged LOS. For each 10-year increase in age, the odds of prolonged LOS increased by 50%. For each 1-point increase in BMI, the odds of prolonged LOS increased by 4%. For each 10-point increase in ODI, the odds of the outcome increased by 20%. For each 20-minute increase in operative time, the odds of prolonged LOS increased by 40%. The ROC curve revealed an AUC of 0.84 (95% CI 0.81–0.86, p < 0.001), indicating excellent discriminatory ability of this predictive model (Appendix 1).

TABLE 4.

Laminectomy group: multivariable logistic regression model for BMI redictive factors of prolonged LOS (> 1 day)

OR95% CIp Value
Patient/clinical variables
 Older age1.051.04–1.07 <0.001
 Living alone1.911.28–2.85 0.02
 BMI1.041.01–1.07 0.018
 Higher ODI1.021.01–1.03 <0.001
Surgical/institutional variables
 Longer op time1.021.01–1.02 <0.001
 Open procedures3.512.51–4.91 <0.001
 Having intra-/postop AEs5.193.45–7.80 <0.001
 No protocol to reduce LOS2.892.05–4.09 <0.001

Boldface type indicates statistical significance.

Posterior Instrumented Fusion Group

The median LOS for the posterior instrumented group was 4 (IQR 2) days. Of the 1639 patients, 1074 (66%) had an LOS less than or equal to 4 days. The fusion group had a range of LOS between 0 and 64 (SD 3.6) days. The frequency distribution of the median LOS is represented in Fig. 2C. Three hundred fifty-four patients (22%) had an LOS above the 75th percentile.

Table 5 displays the sample characteristics, overall and by median. On univariate analysis, patients who experienced prolonged LOS were older, more commonly lived alone, had more comorbidities, less commonly used tobacco, had a lower education level, more commonly used narcotics, more commonly had a chief complaint of neurogenic claudication but less commonly back pain and radiculopathy, and had worse baseline ODI and SF-12 PCS scores (p < 0.05). Patients with prolonged LOS underwent MIS procedures less often, had a longer operative duration and more AEs, and were more likely to be treated at an institution without protocols in place to reduce LOS (p < 0.05).

TABLE 5.

Posterior instrumented fusion group: univariate analysis

Overall (n = 1639)LOS ≤4 Days (n = 1074)LOS >4 Days (n = 565)p Value
Patient/clinical variables
 Male sex, %43.444.142.10.435
 Mean (SD) age, yrs60.1 (12.9)58.2 (12.7)63.8 (12.3) <0.001
 Living arrangement, % (n = 1629)

   Living alone

   Cohabitation


 17.8

 82.2


 14.3

 85.7


 24.5

 75.5
<0.001
 Mean (SD) BMI (n = 1601)29.2 (5.8)29.1 (5.8)29.3 (6.0)0.504
 Mean (SD) no. of comorbidities (n = 1624)3.0 (1.8)2.8 (1.8)3.3 (1.9) <0.001
 Tobacco use, % (n = 1626)18.119.715.2 0.027
 Education ≤ high school, % (n = 1593)42.440.146.7 0.012
 Symptom duration >2 yrs, % (n = 1629)72.872.872.70.959
 Pain medication use, % (n = 1636)88.588.488.70.895
 Narcotic use, % (n = 1410)52.648.959.6 <0.001
 Antidepressant use, % (n = 1392)27.628.425.90.321
 Chief complaint, n (%) (n = 1613)

   Back pain

   Neurogenic claudication

   Radiculopathy

   Other


 269 (16.4)

 691 (42.2)

 647 (39.5)

 6 (0.4)


 201 (18.7)

 403 (37.6)

 457 (42.6)

 2 (0.2)


 68 (12.0)

 288 (51.0)

 190 (33.6)

 4 (0.8)


<0.001

<0.001

<0.001

0.275
 Mean (SD) preop PROs

   NRS back pain (n = 1630)

   NRS leg pain (n = 1631)

   ODI (n = 1631)

   SF-12 MCS (n = 1578)

   SF-12 PCS (n = 1578)

   PHQ-9 (n = 1635)


 7.03 (2.19)

 7.29 (2.17)

 47.0 (14.9)

 44.5 (12.0)

 28.5 (7.5)

 9.45 (6.3)


 7.02 (2.10)

 7.24 (2.16)

 45.6 (14.6)

 44.9 (11.9)

 29.1 (7.4)

 9.3 (6.2)


 7.04 (2.33)

 7.38 (2.21)

 49.6 (15.2)

 43.8 (12.2)

 27.4 (7.7)

 9.7 (6.3)


 0.908

 0.212

<0.001

0.081

<0.001

0.284
Surgical/institutional variables
 MIS, %32.638.421.8 <0.001
 Mean (SD) OR duration, mins (n = 1557)202.8 (90.7)189.3 (88.0)228.2 (90.3) <0.001
 Intra-/postop AEs, %31.518.456.5 <0.001
 Protocol to reduce LOS, % (n = 1581)69.072.961.8 <0.001
 Mean (SD) no. of operated levels1.49 (0.87)1.34 (0.71)1.79 (1.06) <0.001

Percentages are based on the total number of patients in each column. Boldface type indicates statistical significance.

On multivariable analysis, there were 7 predictors of prolonged LOS (Table 6). Those living alone and those being treated at institutions without protocols to reduce LOS had approximately 2 times the odds of prolonged LOS, and those with AEs had close to 5 times the odds of prolonged LOS. For each 10-year increase in age, the odds of prolonged LOS increased by 30%. For each additional comorbidity, the odds of prolonged LOS increased by 9%. For each 2-point decrease in back pain rating, the odds of prolonged LOS increased by 14%. For each 10-point increase in ODI, the odds increased by 20%. For each 20-minute increase in operating room time, the odds of prolonged LOS increased by 8%. The AUC was 0.78 (95% CI 0.75–0.81, p < 0.001), indicating acceptable discriminatory ability of this predictive model (Appendix 1).

TABLE 6.

Posterior instrumented fusion group: multivariable logistic regression model for predictive factors of prolonged LOS (> 4 days)

OR95% CIp Value
Patient/clinical variables
 Older age1.031.02–1.04 <0.001
 Living alone1.491.09–2.03  0.013
 More comorbidities1.091.02–1.17 0.01
 Higher ODI1.021.01–1.03 <0.001
Surgical/institutional variables
 Longer op time1.0041.002–1.005 <0.001
 Having intra-/postop AEs4.713.67–6.04 <0.001
 No protocol to reduce LOS1.711.33–2.21 <0.001

Boldface type indicates statistical significance.

Institutional Practice Survey

Appendix 2 displays a summary of survey results from the 19 institutions. Ten institutions (53%) have either ERAS or other standardized protocols aimed at reducing LOS. Institutions described the elements included in their protocols, which notably included routine outpatient surgery, standardized postoperative clinical care pathways, and routine MIS. Open-ended questions revealed that perceived obstacles to timely discharge were: 1) lack of support at home; 2) lack of access to rehabilitation services; 3) inappropriate patient selection; 4) inadequate pain management; 5) lack of preoperative counseling of patients or contracts on LOS; 6) unrealistic patient expectations; and 7) lack of coordination or a standardized approach between surgeons, nurses, physiotherapists, residents, and fellows. Survey participants suggested that further efforts to shorten LOS should target improving home care and rehabilitation resources and having standardized protocols for discharge, an improved preoperative spine education session, and management of patient expectation as well as preoperative risk assessment for discharge obstacles.

Discussion

This is the first study to build predictive models using individual patient, clinical, surgical, and institutional-level data to investigate the causes of both prolonged LOS and its variability. Common to all groups, worse baseline ODI, experiencing AEs, and being treated at an institution without protocols aimed at reducing LOS were predictive of prolonged LOS. The model performance was acceptable for discectomy and posterior instrumented fusion and excellent for laminectomy.

Specific patient populations are at increased risk for unnecessary delays in discharge, putting them at greater risk for AEs during and after hospitalization.20 Similar to this study, factors such as age, comorbidities, and AEs have previously been reported to be associated with longer LOS.3,6,2123 Comorbidities were predictive of longer LOS for the posterior instrumented fusion group only. Of note, for this analysis, the comorbidities were added and were given equal weight. We also found that AE occurrence was a strong potentially modifiable predictor with a fivefold increase in risk of prolonged LOS, regardless of the procedure. This study reported a high rate of AEs, which is consistent with the use of the SAVES tool, as it captures minor AEs. In an evaluation of the economic impact of perioperative AEs, Hellsten et al.24 reported that minor AEs corresponded to a substantial aggregate cost. This finding is supported by our results, given that LOS is a significant driver of healthcare cost.

Higher baseline ODI scores, reflective of greater disability, were predictive of longer LOS in all surgical groups. Patients with worse preoperative ambulation and functional status have greater needs for rehabilitation and longer recovery time.25 McGirt et al. also reported that patients with higher baseline ODI scores were more likely to have extended LOS following elective lumbar spine surgery.3 Moreover, the current study identified that, in the discectomy group, increased pain severity and longer duration of pain were predictive of increased LOS. Similarly, Virk et al. found that a higher proportion of patients with high preoperative pain scores had a longer LOS after microdiscectomy.26

On multivariable analysis, narcotic usage was not associated with longer LOS in this study. This contrasts with previous studies with a significant association between preoperative narcotic usage and prolonged LOS.2731 Armaghani et al. reported that increased preoperative narcotic use was associated with increased LOS in patients undergoing spine surgery, and with each increase of 100 morphine equivalents, an additional 1.1-day stay was observed.27 It should be noted that narcotic use is a self-reported variable, and it might have been underreported by patients. Also, in our study, this variable was analyzed as a categorical variable (usage or not). The association between opioid use and LOS may be related to the dose and frequency of the narcotic use.

In this study, open surgery was predictive of longer LOS for laminectomy and discectomy. Several previous studies have reported similar findings.3234 Open surgery was not predictive of longer LOS in our patients undergoing posterior instrumented fusion. A hypothesis to explain the lack of association is that this study included fusion up to 5 levels, and the benefit of MIS surgery may be negated. Wang et al. reported a significant reduction in LOS for single-level MIS interbody fusion (3.9 vs 4.8 days, p = 0.017), but not for 2-level MIS interbody fusion surgery (5.1 vs 7.1 days, p = 0.259).35 The benefit of MIS in reducing LOS in patients undergoing posterior instrumented fusion may be dependent on the number of levels fused.

Greater LOS variability may represent opportunities for improvement. In this study, the laminectomy group had the largest range of LOS (0–133 days). The posterior instrumented fusion group had the largest proportion of outliers, with 22% of patients having an LOS above the 75th percentile. This may be explained by the wide spectrum of possible surgeries that this group may have undergone, which creates a heterogeneous group. Preoperative identification of at-risk patients undergoing lumbar surgery, such as the elderly or those living alone, could become a target for LOS reduction with the introduction of mitigation strategies such as an active presurgical discharge planning process. Home care resources become necessary with the growing elderly population undergoing spinal surgery. Adogwa et al. reported a protocol successful in reducing LOS in which elderly patients undergoing lumbar fusion surgery were co-managed by spine surgeons and geriatricians.36 Their protocol reduced LOS by 30% through earlier and improved mobilization and avoidance of postoperative complications.

In the multivariable analysis of all three procedures, being treated at an institution without formal protocols aimed at reducing LOS was predictive of prolonged LOS. Standardized protocols in place included a postoperative clinical care pathway, outpatient surgery, and implementation of MIS. Therefore, creating a protocolized environment with measures targeting timely and safe discharge is a worthwhile strategy to reduce LOS. This premise is also supported by the available literature. Three recent systematic reviews demonstrated the potential of ERAS protocols to reduce LOS with frequent inclusion of preoperative education, discharge planning, the usage of MIS when possible, multimodal analgesia, and early mobilization and rehabilitation.3739 Additionally, Dagal et al. reported a significant reduction in mean hospital LOS for major spine surgery from 8.3 days in traditional care to 6.1 days with enhanced perioperative care.40

The implementation of standardized protocols is an ongoing process, with 10 of the 19 institutions reporting having some of these protocols in place. New ERAS protocols expand to preoperative patient education and patient involvement in discharge planning, which is useful in setting realistic expectations and goals.41 In the recent consensus statement from the ERAS Society for perioperative care in lumbar spine fusion, some factors explored in this study had a strong recommendation grade, such as preoperative education and early mobilization.42 In our survey, close to two-thirds of the participating institutions gave written information about expected LOS and discharge planning preoperatively, but only about one-quarter offered a preoperative spine education session. This could be improved to ensure a better patient understanding of their surgery and postoperative course. With preoperative education, patients have been shown to have better overall satisfaction, especially in terms of their pain management, accelerated functional recovery, decreased morbidity, and better pain scores after lumbar surgery.43 A large meta-analysis showed that patients who received preoperative education spent 1.5 days less in the hospital.44 Prospective research could help provide more robust results on the influence of preoperative patient education and involvement in discharge planning.

While this study provides insight into efforts aimed at reducing LOS, our findings must be interpreted in the context of the study design. The multicenter design and large sample size increases generalizability. Patients enrolled in this study were treated at tertiary and quaternary institutions, and the results might not apply to other clinical settings. This study did not examine in detail the specific components of protocols, which prevents granular insight into which measures included in the protocols are beneficial to decrease LOS. Furthermore, protocols in place vary from one institution to another, and it is possible that patients from an institution with protocols to reduce LOS in place may not have specifically received targeted measures during their hospitalization. Additionally, patients who traveled greater distances to receive spine care might have stayed longer in the hospital, but this variable is not available through the registry. Prospective studies should assess the impact of the various interventions identified in the survey as potential meaningful targets to decrease LOS. Lastly, we observed that high-enrollment institutions were more likely to have protocols in place to reduce LOS. The association between shorter LOS and high-volume hospital and surgeon has been previously reported.45,46 These factors were not included in our analysis, as this study focused more on the implementation of protocols to reduce LOS.

Conclusions

In this first study identifying individual patient, clinical, surgical, and institutional factors across Canada, predictive models were built to enhance the understanding of LOS variability after common elective thoracolumbar spine surgery (discectomy, laminectomy, and posterior instrumented fusion). While many predictive factors differed between the three surgical groups, worse baseline ODI scores, having AEs, and being treated at an institution without protocols aimed at reducing LOS were predictive of prolonged LOS in all groups. The laminectomy group had the largest LOS variability. The current study suggests that institutional protocols such as an ERAS pathway, outpatient surgery, and routine MIS may be effective at decreasing LOS, but further investigation is needed to determine which measures are beneficial.

Disclosures

Dr. Dea: consultant for Stryker, Medtronic, Baxter, and Cerapedics; and direct stock ownership in Medtronic. Dr. Fisher: consultant for Medtronic and NuVasive; and royalties from Medtronic. Dr. Manson: consultant for and support of non–study-related clinical or research from Medtronic. Dr. Phan: consultant for Stryker. Dr. Rampersaud: royalties from Medtronic. Dr. Santaguida: consultant for Symergy and Stryker; and chief medical officer of Careaxis. Dr. Warren: clinical or research support for the study described from Medtronic.

Author Contributions

Conception and design: Dandurand, McIntosh, Charest-Morin. Acquisition of data: Street, Fisher, Finkelstein, Abraham, Paquet, Hall, Wai, Fourney, Bailey, Christie, Soroceanu, Johnson, Kelly, Marion, Nataraj, Santaguida, Warren, Hogan, Manson, Phan, Ahn, Rampersaud, Blanchard, Thomas, Dea, Charest-Morin. Analysis and interpretation of data: Dandurand, McIntosh, Charest-Morin. Drafting the article: Dandurand, Mashayekhi, McIntosh, Charest-Morin. Critically revising the article: Dandurand, McIntosh, Street, Fisher, Finkelstein, Abraham, Paquet, Hall, Wai, Fourney, Bailey, Christie, Soroceanu, Johnson, Kelly, Marion, Nataraj, Santaguida, Warren, Hogan, Manson, Phan, Ahn, Rampersaud, Blanchard, Thomas, Dea, Charest-Morin. Reviewed submitted version of manuscript: Dandurand, McIntosh, Street, Fisher, Finkelstein, Abraham, Paquet, Hall, Wai, Fourney, Bailey, Christie, Soroceanu, Johnson, Kelly, Marion, Nataraj, Santaguida, Warren, Manson, Phan, Ahn, Rampersaud, Blanchard, Thomas, Dea, Charest-Morin. Approved the final version of the manuscript on behalf of all authors: Dandurand. Statistical analysis: McIntosh. Study supervision: Charest-Morin.

Supplemental Information

Online-Only Content

Supplemental material is available with the online version of the article.

Appendices 1–3. https://thejns.org/doi/suppl/10.3171/2022.11.SPINE22662.

References

  • 1

    Rampersaud YR, Power JD, Perruccio AV, et al. Healthcare utilization and costs for spinal conditions in Ontario, Canada—opportunities for funding high-value care: a retrospective cohort study. Spine J. 2020;20(6):874881.

    • PubMed
    • Search Google Scholar
    • Export Citation
  • 2

    Kim RB, Wilkerson C, Karsy M, et al. Prolonged length of stay and risk of unplanned 30-day readmission after elective spine surgery: propensity score-matched analysis of 33,840 patients. Spine (Phila Pa 1976). 2020;45(18):12601268.

    • PubMed
    • Search Google Scholar
    • Export Citation
  • 3

    McGirt MJ, Parker SL, Chotai S, et al. Predictors of extended length of stay, discharge to inpatient rehab, and hospital readmission following elective lumbar spine surgery: introduction of the Carolina-Semmes Grading Scale. J Neurosurg Spine. 2017;27(4):382390.

    • PubMed
    • Search Google Scholar
    • Export Citation
  • 4

    Kobayashi K, Ando K, Kato F, et al. Trends of postoperative length of stay in spine surgery over 10 years in Japan based on a prospective multicenter database. Clin Neurol Neurosurg. 2019;177:97100.

    • PubMed
    • Search Google Scholar
    • Export Citation
  • 5

    Buchbinder R, van Tulder M, Öberg B, et al. Low back pain: a call for action. Lancet. 2018;391(10137):23842388.

  • 6

    Kobayashi K, Ando K, Kato F, et al. Predictors of prolonged length of stay after lumbar interbody fusion: a multicenter study. Global Spine J. 2019;9(5):466472.

    • PubMed
    • Search Google Scholar
    • Export Citation
  • 7

    Halpin RJ, Sugrue PA, Gould RW, et al. Standardizing care for high-risk patients in spine surgery: the Northwestern high-risk spine protocol. Spine (Phila Pa 1976). 2010;35(25):22322238.

    • PubMed
    • Search Google Scholar
    • Export Citation
  • 8

    Fisher CG, Vaccaro AR, Whang PG, et al. Evidence-based recommendations for spine surgery. Spine (Phila Pa 1976). 2013;38(1):E30E37.

  • 9

    Sugrue PA, Halpin RJ, Koski TR. Treatment algorithms and protocol practice in high-risk spine surgery. Neurosurg Clin N Am. 2013;24(2):219230.

  • 10

    Canadian Spine Outcomes and Research Network. Home page. Accessed November 21, 2022. https://www.csorncss.ca/

  • 11

    Childs JD, Piva SR, Fritz JM. Responsiveness of the numeric pain rating scale in patients with low back pain. Spine (Phila Pa 1976). 2005;30(11):13311334.

    • PubMed
    • Search Google Scholar
    • Export Citation
  • 12

    Fairbank JC, Couper J, Davies JB, O’Brien JP. The Oswestry low back pain disability questionnaire. Physiotherapy. 1980;66(8):271273.

  • 13

    Ware J Jr, Kosinski M, Keller SDA. A 12-Item Short-Form Health Survey: construction of scales and preliminary tests of reliability and validity. Med Care. 1996;34(3):220233.

    • PubMed
    • Search Google Scholar
    • Export Citation
  • 14

    Kroenke K, Spitzer RL, Williams JB. The PHQ-9: validity of a brief depression severity measure. J Gen Intern Med. 2001;16(9):606613.

  • 15

    Rampersaud YR, Anderson PA, Dimar JR II, Fisher CG. Spinal Adverse Events Severity System, version 2 (SAVES-V2): inter- and intraobserver reliability assessment. J Neurosurg Spine. 2016;25(2):256263.

    • PubMed
    • Search Google Scholar
    • Export Citation
  • 16

    Collins GS, Reitsma JB, Altman DG, Moons KGM. Transparent reporting of a multivariable prediction model for individual prognosis or diagnosis (TRIPOD): the TRIPOD statement. BMJ. 2015;350:g7594.

    • PubMed
    • Search Google Scholar
    • Export Citation
  • 17

    Collette D. Collett J. Modelling Survival Data in Medical Research. 2nd ed. Taylor & Francis;2003.

  • 18

    Hosmer DW Jr, Lemeshow S, Sturdivant RX. Applied Logistic Regression. 3rd ed. John Wiley & Sons;2013.

  • 19

    Mandrekar JN. Receiver operating characteristic curve in diagnostic test assessment. J Thorac Oncol. 2010;5(9):13151316.

  • 20

    Naessens JM, Campbell CR, Shah N, et al. Effect of illness severity and comorbidity on patient safety and adverse events. Am J Med Qual. 2012;27(1):4857.

  • 21

    Basques BA, Varthi AG, Golinvaux NS, Bohl DD, Grauer JN. Patient characteristics associated with increased postoperative length of stay and readmission after elective laminectomy for lumbar spinal stenosis. Spine (Phila Pa 1976). 2014;39(10):833840.

    • PubMed
    • Search Google Scholar
    • Export Citation
  • 22

    Shin JIB, Kim JS, Steinberger J, DiCapua J, Cho SK. Patient factors contributing to prolonged postoperative length of stay and increased rate of readmission after elective posterior cervical fusion. Clin Spine Surg. 2018;31(1):E55E61.

    • PubMed
    • Search Google Scholar
    • Export Citation
  • 23

    Zheng F, Cammisa FPJ Jr, Sandhu HS, Girardi FP, Khan SN. Factors predicting hospital stay, operative time, blood loss, and transfusion in patients undergoing revision posterior lumbar spine decompression, fusion, and segmental instrumentation. Spine (Phila Pa 1976). 2002;27(8):818824.

    • PubMed
    • Search Google Scholar
    • Export Citation
  • 24

    Hellsten EK, Hanbidge MA, Manos AN, et al. An economic evaluation of perioperative adverse events associated with spinal surgery. Spine J. 2013;13(1):4453.

  • 25

    Kanaan SF, Yeh HW, Waitman RL, Burton DC, Arnold PM, Sharma NK. Predicting discharge placement and health care needs after lumbar spine laminectomy. J Allied Health. 2014;43(2):8897.

    • PubMed
    • Search Google Scholar
    • Export Citation
  • 26

    Virk S, Vaishnav AS, Mok JK, et al. How do high preoperative pain scores impact the clinical course and outcomes for patients undergoing lumbar microdiscectomy? J Neurosurg Spine. 2020;33(6):772778.

    • PubMed
    • Search Google Scholar
    • Export Citation
  • 27

    Armaghani SJ, Lee DS, Bible JE, et al. Increased preoperative narcotic use and its association with postoperative complications and length of hospital stay in patients undergoing spine surgery. Clin Spine Surg. 2016;29(2):E93E98.

    • PubMed
    • Search Google Scholar
    • Export Citation
  • 28

    Li Y, Stocchi L, Cherla D, Liu X, Remzi FH. Association of preoperative narcotic use with postoperative complications and prolonged length of hospital stay in patients with crohn disease. JAMA Surg. 2016;151(8):726734.

    • PubMed
    • Search Google Scholar
    • Export Citation
  • 29

    Buckarma E, Thiels CA, Habermann EB, et al. Preoperative opioid use is associated with increased length of stay after pancreaticoduodenectomy. HPB (Oxford). 2020;22(7):10741081.

    • PubMed
    • Search Google Scholar
    • Export Citation
  • 30

    Ogilvie JW Jr, Wilkes AW, Hobbs DJ, Smith JR, Dull MB, Luchtefeld MA. The effect of chronic preoperative opioid use on surgical site infections, length of stay, and readmissions. Dis Colon Rectum. 2020;63(9):13101316.

    • PubMed
    • Search Google Scholar
    • Export Citation
  • 31

    Doan LV, Wang J, Padjen K, et al. Preoperative long-acting opioid use is associated with increased length of stay and readmission rates after elective surgeries. Pain Med. 2019;20(12):25392551.

    • PubMed
    • Search Google Scholar
    • Export Citation
  • 32

    Lundberg J, Langevin JP. Lumbar microlaminectomy vs traditional laminectomy. Fed Pract. 2017;34(12):3235.

  • 33

    Phan K, Mobbs RJ. Minimally invasive versus open laminectomy for lumbar stenosis: a systematic review and Meta-analysis. Spine (Phila Pa 1976). 2016;41(2):E91E100.

    • PubMed
    • Search Google Scholar
    • Export Citation
  • 34

    Evaniew N, Bogle A, Soroceanu A, et al. Minimally invasive tubular lumbar discectomy versus conventional open lumbar discectomy: an observational study from the Canadian Spine Outcomes and Research Network. Global Spine J. Published online July 9, 2021. doi: 10.1177/21925682211029863

    • PubMed
    • Search Google Scholar
    • Export Citation
  • 35

    Wang MY, Cummock MD, Yu Y, Trivedi RA. An analysis of the differences in the acute hospitalization charges following minimally invasive versus open posterior lumbar interbody fusion. J Neurosurg Spine. 2010;12(6):694699.

    • PubMed
    • Search Google Scholar
    • Export Citation
  • 36

    Adogwa O, Elsamadicy AA, Vuong VD, et al. Geriatric comanagement reduces perioperative complications and shortens duration of hospital stay after lumbar spine surgery: a prospective single-institution experience. J Neurosurg Spine. 2017;27(6):670675.

    • PubMed
    • Search Google Scholar
    • Export Citation
  • 37

    Elsarrag M, Soldozy S, Patel P, et al. Enhanced recovery after spine surgery: a systematic review. Neurosurg Focus. 2019;46(4):E3.

  • 38

    Tong Y, Fernandez L, Bendo JA, Spivak JM. Enhanced Recovery After Surgery trends in adult spine surgery: a systematic review. Int J Spine Surg. 2020;14(4):623640.

    • PubMed
    • Search Google Scholar
    • Export Citation
  • 39

    Licina A, Silvers A, Laughlin H, Russell J, Wan C. Pathway for enhanced recovery after spinal surgery—a systematic review of evidence for use of individual components. BMC Anesthesiol. 2021;21(1):74.

    • PubMed
    • Search Google Scholar
    • Export Citation
  • 40

    Dagal A, Bellabarba C, Bransford R, et al. Enhanced perioperative care for major spine surgery. Spine (Phila Pa 1976). 2019;44(13):959966.

  • 41

    Carr DA, Saigal R, Zhang F, Bransford RJ, Bellabarba C, Dagal A. Enhanced perioperative care and decreased cost and length of stay after elective major spinal surgery. Neurosurg Focus. 2019;46(4):E5.

    • PubMed
    • Search Google Scholar
    • Export Citation
  • 42

    Debono B, Wainwright TW, Wang MY, et al. Consensus statement for perioperative care in lumbar spinal fusion: Enhanced Recovery After Surgery (ERAS®) Society recommendations. Spine J. 2021;21(5):729752.

    • PubMed
    • Search Google Scholar
    • Export Citation
  • 43

    Papanastassiou I, Anderson R, Barber N, Conover C, Castellvi AE. Effects of preoperative education on spinal surgery patients. SAS J. 2011;5(4):120124.

    • PubMed
    • Search Google Scholar
    • Export Citation
  • 44

    Devine EC. Effects of psychoeducational care for adult surgical patients: a meta-analysis of 191 studies. Patient Educ Couns. 1992;19(2):129142.

  • 45

    Adkins ZB, Malik AT, Jain N, Yu E, Kim J, Khan SN. Does hospital volume affect outcomes in spine surgeries? A systematic review. Clin Spine Surg. 2019;32(7):285294.

    • PubMed
    • Search Google Scholar
    • Export Citation
  • 46

    Malik AT, Panni UY, Mirza MU, Tetlay M, Noordin S. The impact of surgeon volume on patient outcome in spine surgery: a systematic review. Eur Spine J. 2018;27(3):530542.

    • PubMed
    • Search Google Scholar
    • Export Citation

Supplementary Materials

  • Collapse
  • Expand

Illustration from Akinduro et al. (pp 481–493). Used with permission of Mayo Foundation for Medical Education and Research, all rights reserved.

  • FIG. 1.

    Flowchart of patient inclusion and exclusion. TL = thoracolumbar. *Patients were analyzed in the laminectomy group.

  • FIG. 2.

    Frequency distribution of the median LOS per procedure. A: Discectomy. B: Laminectomy. C: Posterior thoracolumbar fusion.

  • 1

    Rampersaud YR, Power JD, Perruccio AV, et al. Healthcare utilization and costs for spinal conditions in Ontario, Canada—opportunities for funding high-value care: a retrospective cohort study. Spine J. 2020;20(6):874881.

    • PubMed
    • Search Google Scholar
    • Export Citation
  • 2

    Kim RB, Wilkerson C, Karsy M, et al. Prolonged length of stay and risk of unplanned 30-day readmission after elective spine surgery: propensity score-matched analysis of 33,840 patients. Spine (Phila Pa 1976). 2020;45(18):12601268.

    • PubMed
    • Search Google Scholar
    • Export Citation
  • 3

    McGirt MJ, Parker SL, Chotai S, et al. Predictors of extended length of stay, discharge to inpatient rehab, and hospital readmission following elective lumbar spine surgery: introduction of the Carolina-Semmes Grading Scale. J Neurosurg Spine. 2017;27(4):382390.

    • PubMed
    • Search Google Scholar
    • Export Citation
  • 4

    Kobayashi K, Ando K, Kato F, et al. Trends of postoperative length of stay in spine surgery over 10 years in Japan based on a prospective multicenter database. Clin Neurol Neurosurg. 2019;177:97100.

    • PubMed
    • Search Google Scholar
    • Export Citation
  • 5

    Buchbinder R, van Tulder M, Öberg B, et al. Low back pain: a call for action. Lancet. 2018;391(10137):23842388.

  • 6

    Kobayashi K, Ando K, Kato F, et al. Predictors of prolonged length of stay after lumbar interbody fusion: a multicenter study. Global Spine J. 2019;9(5):466472.

    • PubMed
    • Search Google Scholar
    • Export Citation
  • 7

    Halpin RJ, Sugrue PA, Gould RW, et al. Standardizing care for high-risk patients in spine surgery: the Northwestern high-risk spine protocol. Spine (Phila Pa 1976). 2010;35(25):22322238.

    • PubMed
    • Search Google Scholar
    • Export Citation
  • 8

    Fisher CG, Vaccaro AR, Whang PG, et al. Evidence-based recommendations for spine surgery. Spine (Phila Pa 1976). 2013;38(1):E30E37.

  • 9

    Sugrue PA, Halpin RJ, Koski TR. Treatment algorithms and protocol practice in high-risk spine surgery. Neurosurg Clin N Am. 2013;24(2):219230.

  • 10

    Canadian Spine Outcomes and Research Network. Home page. Accessed November 21, 2022. https://www.csorncss.ca/

  • 11

    Childs JD, Piva SR, Fritz JM. Responsiveness of the numeric pain rating scale in patients with low back pain. Spine (Phila Pa 1976). 2005;30(11):13311334.

    • PubMed
    • Search Google Scholar
    • Export Citation
  • 12

    Fairbank JC, Couper J, Davies JB, O’Brien JP. The Oswestry low back pain disability questionnaire. Physiotherapy. 1980;66(8):271273.

  • 13

    Ware J Jr, Kosinski M, Keller SDA. A 12-Item Short-Form Health Survey: construction of scales and preliminary tests of reliability and validity. Med Care. 1996;34(3):220233.

    • PubMed
    • Search Google Scholar
    • Export Citation
  • 14

    Kroenke K, Spitzer RL, Williams JB. The PHQ-9: validity of a brief depression severity measure. J Gen Intern Med. 2001;16(9):606613.

  • 15

    Rampersaud YR, Anderson PA, Dimar JR II, Fisher CG. Spinal Adverse Events Severity System, version 2 (SAVES-V2): inter- and intraobserver reliability assessment. J Neurosurg Spine. 2016;25(2):256263.

    • PubMed
    • Search Google Scholar
    • Export Citation
  • 16

    Collins GS, Reitsma JB, Altman DG, Moons KGM. Transparent reporting of a multivariable prediction model for individual prognosis or diagnosis (TRIPOD): the TRIPOD statement. BMJ. 2015;350:g7594.

    • PubMed
    • Search Google Scholar
    • Export Citation
  • 17

    Collette D. Collett J. Modelling Survival Data in Medical Research. 2nd ed. Taylor & Francis;2003.

  • 18

    Hosmer DW Jr, Lemeshow S, Sturdivant RX. Applied Logistic Regression. 3rd ed. John Wiley & Sons;2013.

  • 19

    Mandrekar JN. Receiver operating characteristic curve in diagnostic test assessment. J Thorac Oncol. 2010;5(9):13151316.

  • 20

    Naessens JM, Campbell CR, Shah N, et al. Effect of illness severity and comorbidity on patient safety and adverse events. Am J Med Qual. 2012;27(1):4857.

  • 21

    Basques BA, Varthi AG, Golinvaux NS, Bohl DD, Grauer JN. Patient characteristics associated with increased postoperative length of stay and readmission after elective laminectomy for lumbar spinal stenosis. Spine (Phila Pa 1976). 2014;39(10):833840.

    • PubMed
    • Search Google Scholar
    • Export Citation
  • 22

    Shin JIB, Kim JS, Steinberger J, DiCapua J, Cho SK. Patient factors contributing to prolonged postoperative length of stay and increased rate of readmission after elective posterior cervical fusion. Clin Spine Surg. 2018;31(1):E55E61.

    • PubMed
    • Search Google Scholar
    • Export Citation
  • 23

    Zheng F, Cammisa FPJ Jr, Sandhu HS, Girardi FP, Khan SN. Factors predicting hospital stay, operative time, blood loss, and transfusion in patients undergoing revision posterior lumbar spine decompression, fusion, and segmental instrumentation. Spine (Phila Pa 1976). 2002;27(8):818824.

    • PubMed
    • Search Google Scholar
    • Export Citation
  • 24

    Hellsten EK, Hanbidge MA, Manos AN, et al. An economic evaluation of perioperative adverse events associated with spinal surgery. Spine J. 2013;13(1):4453.

  • 25

    Kanaan SF, Yeh HW, Waitman RL, Burton DC, Arnold PM, Sharma NK. Predicting discharge placement and health care needs after lumbar spine laminectomy. J Allied Health. 2014;43(2):8897.

    • PubMed
    • Search Google Scholar
    • Export Citation
  • 26

    Virk S, Vaishnav AS, Mok JK, et al. How do high preoperative pain scores impact the clinical course and outcomes for patients undergoing lumbar microdiscectomy? J Neurosurg Spine. 2020;33(6):772778.

    • PubMed
    • Search Google Scholar
    • Export Citation
  • 27

    Armaghani SJ, Lee DS, Bible JE, et al. Increased preoperative narcotic use and its association with postoperative complications and length of hospital stay in patients undergoing spine surgery. Clin Spine Surg. 2016;29(2):E93E98.

    • PubMed
    • Search Google Scholar
    • Export Citation
  • 28

    Li Y, Stocchi L, Cherla D, Liu X, Remzi FH. Association of preoperative narcotic use with postoperative complications and prolonged length of hospital stay in patients with crohn disease. JAMA Surg. 2016;151(8):726734.

    • PubMed
    • Search Google Scholar
    • Export Citation
  • 29

    Buckarma E, Thiels CA, Habermann EB, et al. Preoperative opioid use is associated with increased length of stay after pancreaticoduodenectomy. HPB (Oxford). 2020;22(7):10741081.

    • PubMed
    • Search Google Scholar
    • Export Citation
  • 30

    Ogilvie JW Jr, Wilkes AW, Hobbs DJ, Smith JR, Dull MB, Luchtefeld MA. The effect of chronic preoperative opioid use on surgical site infections, length of stay, and readmissions. Dis Colon Rectum. 2020;63(9):13101316.

    • PubMed
    • Search Google Scholar
    • Export Citation
  • 31

    Doan LV, Wang J, Padjen K, et al. Preoperative long-acting opioid use is associated with increased length of stay and readmission rates after elective surgeries. Pain Med. 2019;20(12):25392551.

    • PubMed
    • Search Google Scholar
    • Export Citation
  • 32

    Lundberg J, Langevin JP. Lumbar microlaminectomy vs traditional laminectomy. Fed Pract. 2017;34(12):3235.

  • 33

    Phan K, Mobbs RJ. Minimally invasive versus open laminectomy for lumbar stenosis: a systematic review and Meta-analysis. Spine (Phila Pa 1976). 2016;41(2):E91E100.

    • PubMed
    • Search Google Scholar
    • Export Citation
  • 34

    Evaniew N, Bogle A, Soroceanu A, et al. Minimally invasive tubular lumbar discectomy versus conventional open lumbar discectomy: an observational study from the Canadian Spine Outcomes and Research Network. Global Spine J. Published online July 9, 2021. doi: 10.1177/21925682211029863

    • PubMed
    • Search Google Scholar
    • Export Citation
  • 35

    Wang MY, Cummock MD, Yu Y, Trivedi RA. An analysis of the differences in the acute hospitalization charges following minimally invasive versus open posterior lumbar interbody fusion. J Neurosurg Spine. 2010;12(6):694699.

    • PubMed
    • Search Google Scholar
    • Export Citation
  • 36

    Adogwa O, Elsamadicy AA, Vuong VD, et al. Geriatric comanagement reduces perioperative complications and shortens duration of hospital stay after lumbar spine surgery: a prospective single-institution experience. J Neurosurg Spine. 2017;27(6):670675.

    • PubMed
    • Search Google Scholar
    • Export Citation
  • 37

    Elsarrag M, Soldozy S, Patel P, et al. Enhanced recovery after spine surgery: a systematic review. Neurosurg Focus. 2019;46(4):E3.

  • 38

    Tong Y, Fernandez L, Bendo JA, Spivak JM. Enhanced Recovery After Surgery trends in adult spine surgery: a systematic review. Int J Spine Surg. 2020;14(4):623640.

    • PubMed
    • Search Google Scholar
    • Export Citation
  • 39

    Licina A, Silvers A, Laughlin H, Russell J, Wan C. Pathway for enhanced recovery after spinal surgery—a systematic review of evidence for use of individual components. BMC Anesthesiol. 2021;21(1):74.

    • PubMed
    • Search Google Scholar
    • Export Citation
  • 40

    Dagal A, Bellabarba C, Bransford R, et al. Enhanced perioperative care for major spine surgery. Spine (Phila Pa 1976). 2019;44(13):959966.

  • 41

    Carr DA, Saigal R, Zhang F, Bransford RJ, Bellabarba C, Dagal A. Enhanced perioperative care and decreased cost and length of stay after elective major spinal surgery. Neurosurg Focus. 2019;46(4):E5.

    • PubMed
    • Search Google Scholar
    • Export Citation
  • 42

    Debono B, Wainwright TW, Wang MY, et al. Consensus statement for perioperative care in lumbar spinal fusion: Enhanced Recovery After Surgery (ERAS®) Society recommendations. Spine J. 2021;21(5):729752.

    • PubMed
    • Search Google Scholar
    • Export Citation
  • 43

    Papanastassiou I, Anderson R, Barber N, Conover C, Castellvi AE. Effects of preoperative education on spinal surgery patients. SAS J. 2011;5(4):120124.

    • PubMed
    • Search Google Scholar
    • Export Citation
  • 44

    Devine EC. Effects of psychoeducational care for adult surgical patients: a meta-analysis of 191 studies. Patient Educ Couns. 1992;19(2):129142.

  • 45

    Adkins ZB, Malik AT, Jain N, Yu E, Kim J, Khan SN. Does hospital volume affect outcomes in spine surgeries? A systematic review. Clin Spine Surg. 2019;32(7):285294.

    • PubMed
    • Search Google Scholar
    • Export Citation
  • 46

    Malik AT, Panni UY, Mirza MU, Tetlay M, Noordin S. The impact of surgeon volume on patient outcome in spine surgery: a systematic review. Eur Spine J. 2018;27(3):530542.

    • PubMed
    • Search Google Scholar
    • Export Citation

Metrics

All Time Past Year Past 30 Days
Abstract Views 1453 324 0
Full Text Views 719 193 41
PDF Downloads 542 224 18
EPUB Downloads 0 0 0