Minimally invasive versus open lumbar spinal fusion: a matched study investigating patient-reported and surgical outcomes

View More View Less
  • 1 Department of Neurosurgery, University of Alabama at Birmingham, Birmingham, Alabama;
  • | 2 Mayo Clinic Neuro-Informatics Laboratory, Department of Neurologic Surgery, Mayo Clinic, Rochester, Minnesota;
  • | 3 Department of Neurologic Surgery, Mayo Clinic, Rochester, Minnesota;
  • | 4 School of Medicine, University of North Carolina at Chapel Hill, Chapel Hill, North Carolina;
  • | 5 Department of Neurological Surgery, University of California, San Francisco, California;
  • | 6 Department of Neurosurgery, Clinical Neurosciences Center, University of Utah, Salt Lake City, Utah;
  • | 7 Atlanta Brain and Spine Care, Atlanta, Georgia;
  • | 8 Atlantic Neurosurgical Specialists, Morristown, New Jersey;
  • | 9 Steamboat Orthopaedic and Spine Institute, Steamboat Springs, Colorado;
  • | 10 Department of Neurosurgery, Washington University School of Medicine, St. Louis, Missouri; and
  • | 11 Neuroscience Institute, Carolina Neurosurgery & Spine Associates, Carolinas Healthcare System, Charlotte, North Carolina
Free access

OBJECTIVE

With the expanding indications for and increasing popularity of minimally invasive surgery (MIS) for lumbar spinal fusion, large-scale outcomes analysis to compare MIS approaches with open procedures is warranted.

METHODS

The authors queried the Quality Outcomes Database for patients who underwent elective lumbar fusion for degenerative spine disease. They performed optimal matching, at a 1:2 ratio between patients who underwent MIS and those who underwent open lumbar fusion, to create two highly homogeneous groups in terms of 33 baseline variables (including demographic characteristics, comorbidities, symptoms, patient-reported scores, indications, and operative details). The outcomes of interest were overall satisfaction, decrease in Oswestry Disability Index (ODI), and back and leg pain, as well as hospital length of stay (LOS), operative time, reoperations, and incidental durotomy rate. Satisfaction was defined as a score of 1 or 2 on the North American Spine Society scale. Minimal clinically important difference (MCID) in ODI was defined as ≥ 30% decrease from baseline. Outcomes were assessed at the 3- and 12-month follow-up evaluations.

RESULTS

After the groups were matched, the MIS and open groups consisted of 1483 and 2966 patients, respectively. Patients who underwent MIS fusion had higher odds of satisfaction at 3 months (OR 1.4, p = 0.004); no difference was demonstrated at 12 months (OR 1.04, p = 0.67). Lumbar stenosis, single-level fusion, higher American Society of Anesthesiologists Physical Status Classification System grade, and absence of spondylolisthesis were most prominently associated with higher odds of satisfaction with MIS compared with open surgery. Patients in the MIS group had slightly lower ODI scores at 3 months (mean difference 1.61, p = 0.006; MCID OR 1.14, p = 0.0495) and 12 months (mean difference 2.35, p < 0.001; MCID OR 1.29, p < 0.001). MIS was also associated with a greater decrease in leg and back pain at both follow-up time points. The two groups did not differ in operative time and incidental durotomy rate; however, LOS was shorter for the MIS group. Revision surgery at 12 months was less likely for patients who underwent MIS (4.1% vs 5.6%, p = 0.032).

CONCLUSIONS

In patients who underwent lumbar fusion for degenerative spinal disease, MIS was associated with higher odds of satisfaction at 3 months postoperatively. No difference was demonstrated at the 12-month follow-up. MIS maintained a small, yet consistent, superiority in decreasing ODI and back and leg pain, and MIS was associated with a lower reoperation rate.

ABBREVIATIONS

ASA = American Society of Anesthesiologists Physical Status Classification System; LOS = length of stay; MCID = minimal clinically important difference; MIS = minimally invasive surgery; NASS = North American Spine Society; NNT = number needed to treat; ODI = Oswestry Disability Index; PRO = patient-reported outcome; QOD = Quality Outcomes Database; SMD = standardized mean difference; VAS = visual analog scale.

OBJECTIVE

With the expanding indications for and increasing popularity of minimally invasive surgery (MIS) for lumbar spinal fusion, large-scale outcomes analysis to compare MIS approaches with open procedures is warranted.

METHODS

The authors queried the Quality Outcomes Database for patients who underwent elective lumbar fusion for degenerative spine disease. They performed optimal matching, at a 1:2 ratio between patients who underwent MIS and those who underwent open lumbar fusion, to create two highly homogeneous groups in terms of 33 baseline variables (including demographic characteristics, comorbidities, symptoms, patient-reported scores, indications, and operative details). The outcomes of interest were overall satisfaction, decrease in Oswestry Disability Index (ODI), and back and leg pain, as well as hospital length of stay (LOS), operative time, reoperations, and incidental durotomy rate. Satisfaction was defined as a score of 1 or 2 on the North American Spine Society scale. Minimal clinically important difference (MCID) in ODI was defined as ≥ 30% decrease from baseline. Outcomes were assessed at the 3- and 12-month follow-up evaluations.

RESULTS

After the groups were matched, the MIS and open groups consisted of 1483 and 2966 patients, respectively. Patients who underwent MIS fusion had higher odds of satisfaction at 3 months (OR 1.4, p = 0.004); no difference was demonstrated at 12 months (OR 1.04, p = 0.67). Lumbar stenosis, single-level fusion, higher American Society of Anesthesiologists Physical Status Classification System grade, and absence of spondylolisthesis were most prominently associated with higher odds of satisfaction with MIS compared with open surgery. Patients in the MIS group had slightly lower ODI scores at 3 months (mean difference 1.61, p = 0.006; MCID OR 1.14, p = 0.0495) and 12 months (mean difference 2.35, p < 0.001; MCID OR 1.29, p < 0.001). MIS was also associated with a greater decrease in leg and back pain at both follow-up time points. The two groups did not differ in operative time and incidental durotomy rate; however, LOS was shorter for the MIS group. Revision surgery at 12 months was less likely for patients who underwent MIS (4.1% vs 5.6%, p = 0.032).

CONCLUSIONS

In patients who underwent lumbar fusion for degenerative spinal disease, MIS was associated with higher odds of satisfaction at 3 months postoperatively. No difference was demonstrated at the 12-month follow-up. MIS maintained a small, yet consistent, superiority in decreasing ODI and back and leg pain, and MIS was associated with a lower reoperation rate.

In Brief

The authors compared minimally invasive surgery (MIS) and open spinal fusion for degenerative spine disease in terms of clinical and patient-reported outcomes 1 year after surgery. Optimal matching for 33 baseline characteristics was performed to create two highly homogeneous groups. MIS was associated with decreased duration of hospitalization, higher odds of patient satisfaction at 3 months, and fewer reoperations at 12 months. This study provides evidence of the advantages of MIS for lumbar spinal fusion.

Lumbar spinal fusion is commonly used to stabilize patients with degenerative disease, spondylolisthesis, lumbar stenosis, and other spine conditions. These surgical procedures have traditionally been performed using open techniques; however, open spine surgery involves increased trauma to the muscle and soft tissues of the spine, leading to prolonged recovery times. Minimally invasive surgery (MIS) techniques allow removal of tissue from the surgeon’s visual field without the need to make a large exposing incision, thereby sparing surrounding tissue.

MIS for lumbar spinal fusion reduces trauma to adjacent tissue and is associated with a reduced risk of muscle damage, less blood loss, and shorter hospital length of stay (LOS).15 Although decreased blood loss and reduced postoperative pain have been associated with MIS, some studies have demonstrated no difference in clinical outcomes at 12 months, and others have demonstrated increased patient satisfaction with open decompression of low-grade spondylolisthesis.3,4 Another study found no difference between the two approaches for 1-level fusion, with a greater decrease in leg pain in the MIS group.6

As MIS techniques become more popular, it is important to measure the success of these procedures by utilizing patient-reported outcomes (PROs). Large-scale studies that compare the outcomes of MIS approaches to those of traditional open procedures have not been performed. The primary objective of this study was to compare MIS and open surgery for lumbar spinal fusion by utilizing PROs from the lumbar spine surgery module of the Quality Outcomes Database (QOD).

Methods

The Quality Outcomes Database

The lumbar spine surgery module of the QOD registry was queried for this project. The QOD is a nationwide registry that was instituted in 2012 by the NeuroPoint Alliance. The QOD has impacted surgical outcomes research in the field of spine surgery mainly because of three unique features. First, the QOD is unique among databases in terms of scale and granularity. Second, the QOD includes PROs, such as pain on the visual analog scale (VAS), the Oswestry Disability Index (ODI), and patient satisfaction on the 4-point North American Spine Society (NASS) scale. These outcomes are of particular importance in spine surgery because they illustrate the primary aim of the intervention, namely to alleviate symptoms and improve quality of life.7 Third, the QOD follows patients longitudinally, thereby providing insights into the long-term efficacy of surgical interventions.

Patient Cohort

The lumbar spine surgery module of the QOD contains data for more than 70,000 patients from 220 participating sites. For this project, only patients who underwent spinal fusion for degenerative indications with available 1-year follow-up data were included. Patients who underwent lumbar fusion via a posterior approach were included in the MIS group if surgery was performed with a tubular retractor. For anterior or lateral interbody fusions, surgery was considered MIS if screw fixation was performed percutaneously. The remaining patients were included in the open group.

Outcomes of Interest

The primary endpoints of this study were patient satisfaction and decrease from baseline in ODI at 3 months after surgery. Patient satisfaction was measured on the basis of the NASS scale; patients were considered satisfied if they had a NASS score of 1 or 2.

The secondary endpoints of this study included LOS, operative time, rate of incidental durotomy, readmission rate within 30 days, revision surgery rate at 3 and 12 months, and satisfaction and decrease in ODI at 12 months, as well as VAS scores for back pain and leg pain at 3 and 12 months. Minimal clinically important difference (MCID) in ODI and VAS scores was defined as a 30% decrease from baseline.8 Optimal ODI at follow-up was defined as < 20, optimal VAS score for pain was defined as ≤ 2, and maximal satisfaction was defined as a NASS score equal to 1.

Statistical Analysis

Continuous data are presented as mean ± SD, and categorical data are presented as frequencies and proportions. The unpaired 2-sample t-test was performed for univariate analysis of continuous data, and Pearson’s chi-square test was performed for categorical data. Multivariable logistic regression was performed to identify characteristics independently associated with the decision to perform either MIS or open lumbar fusion. The R language and RStudio environment (version 4.0.5) were used for all statistical analyses.9 The results were considered statistically significant for p < 0.05. The rms package was utilized for logistic regression.

Optimal Matching

To create two highly homogeneous groups, we matched each patient in the MIS group with 2 patients in the open group via optimal matching. Optimal matching belongs to the family of matching techniques (which also includes propensity score and greedy matching) that are used to simulate randomization in observational studies. Optimal matching differs from propensity score matching in that patients are matched directly on the basis of their characteristics instead of the propensity score (the probability of receiving one treatment over the other) that these characteristics have.10 Optimal matching is preferable to greedy matching because it pursues greater homogeneity of the total cohort instead of serially finding the ideal match (i.e., nearest neighbor) for individual patients.11

In randomized clinical trials, patients have the same probability of being allocated to each group. This feature guarantees that there is no selection bias and that the two arms differ only in terms of the treatment received; therefore, any difference in outcomes can be safely attributed to the effect of treatment. In observational studies, this hypothesis of equal probability of receiving each treatment can be approximated by utilizing matching techniques.12 The rationale is that if the two groups contain patients who are highly similar in terms of the baseline characteristics that influence decision-making and postoperative outcomes, then one could safely assume that the differences in outcomes between the two groups are not products of confounders but instead due to differences in the treatment methods. Hence, the number and clinical significance of the variables used to match the two groups are critical for valid comparison.

In this study, we matched the MIS and open groups in terms of 33 baseline variables via optimal matching. The cutoff value of the standardized mean difference (SMD) for sufficient matching was set at less than 0.1.13 A matching ratio of 1:2 was selected to increase the power of the study after making sure that the cutoff criteria for sufficient matching were not breached. The MatchIt package was utilized for the statistical analysis of optimal matching.

Results

Study Cohort

The lumbar module of the QOD included 72,504 patients. A total of 3363 patients underwent surgery for correction of deformity and were, therefore, beyond the scope of this study. Of the remaining 69,141 patients, 26,393 underwent procedures that included arthrodesis; the rest were excluded from further analysis. Furthermore, 543 of 2343 patients who underwent 2-stage surgery were excluded because one of the stages was MIS and the other was open surgery. Of 25,850 eligible patients, 19,921 patients (77%) were followed up for 3 months and 13,990 patients (54.1%) were followed up for 12 months. Finally, 11,213 patients (43.4%) with available data on all baseline and operative covariates of interest, as well as all PROs of interest at both time points, were included in the analysis.

The final patient group consisted of 11,213 patients included in the lumbar spine surgery module of the QOD. Of these patients, 9730 underwent open lumbar fusion and 1483 underwent MIS lumbar fusion for degenerative indications. The baseline characteristics of these groups are available in Table 1.

TABLE 1.

Baseline characteristics of the patients who underwent MIS and open lumbar fusion prior to matching

CharacteristicOpen (n = 9730)MIS (n = 1483)Total (n = 11,213)p ValueSMD
Age, yrs62.419 (11.678)61.652 (11.627)62.318 (11.674)0.0180.066
BMI, kg/m230.933 (6.222)30.541 (6.225)30.881 (6.224)0.0240.063
Smoker1245 (12.8)162 (10.9)1407 (12.5)0.0430.058
Female sex5316 (54.6)801 (54.0)6117 (54.6)0.6540.013
Revision1765 (18.1)180 (12.1)1945 (17.3)<0.0010.168
Osteoporosis626 (6.4)84 (5.7)710 (6.3)0.2570.032
Diabetes mellitus2045 (21.0)256 (17.3)2301 (20.5)<0.0010.096
ASA grade<0.0010.183
 I188 (1.9)52 (3.5)240 (2.1)
 II4618 (47.5)798 (53.8)5416 (48.3)
 III4791 (49.2)625 (42.1)5416 (48.3)
 IV133 (1.4)8 (0.5)141 (1.3)
Ambulation<0.0010.154
 Independent8162 (84.0)1327 (89.5)9489 (84.7)
 Partially dependent1492 (15.3)147 (9.9)1639 (14.6)
 Totally dependent67 (0.7)9 (0.6)76 (0.7)
Education level<0.0010.100
 Not available257 (2.6)24 (1.6)281 (2.5)
 High school or less4744 (48.8)652 (44.0)5396 (48.1)
 Graduate level3473 (35.7)608 (41.0)4081 (36.4)
 Postgraduate level1256 (12.9)199 (13.4)1455 (13.0)
Symptomatic disc herniation1702 (17.5)287 (19.4)1989 (17.7)0.0810.048
Grade I spondylolisthesis5270 (54.2)875 (59.0)6145 (54.8)<0.0010.098
Lumbar stenosis5540 (56.9)813 (54.8)6353 (56.7)0.1260.043
Adjacent-segment disease1148 (11.8)89 (6.0)1237 (11.0)<0.0010.205
Single-level symptomatic mechanical disc collapse222 (2.3)12 (0.8)234 (2.1)<0.0010.120
Pseudarthrosis219 (2.3)6 (0.4)225 (2.0)<0.0010.162
ODI at baseline46.570 (15.904)45.061 (15.771)46.371 (15.894)<0.0010.095
VAS
 Leg pain6.613 (2.781)6.567 (2.784)6.607 (2.781)0.5540.016
 Back pain6.971 (2.450)6.764 (2.604)6.944 (2.471)0.0030.082
Race
 African American604 (6.2)112 (7.6)716 (6.4)0.0490.053
 White8711 (89.5)1317 (88.8)10,028 (89.4)0.4000.023
 Other230 (2.4)34 (2.3)264 (2.4)0.8660.005
Insurance<0.0010.206
 Medicare4624 (47.5)617 (41.6)5241 (46.7)
 Medicaid409 (4.2)33 (2.2)442 (3.9)
 Private4306 (44.3)791 (53.3)5097 (45.5)
 Uninsured61 (0.6)4 (0.3)65 (0.6)
 Veterans Affairs/government330 (3.4)38 (2.6)368 (3.3)
Workers’ compensation272 (2.8)23 (1.6)295 (2.6)0.0050.085
Liability claim437 (4.5)43 (2.9)480 (4.3)0.0050.084
No. of levels fused1.441 (0.639)1.220 (0.471)1.412 (0.624)<0.0010.394
Levels fused<0.0010.394
 16224 (64.0)1194 (80.5)7418 (66.2)
 22721 (28.0)252 (17.0)2973 (26.5)
 3785 (8.1)37 (2.5)822 (7.3)
Interbody graft8834 (90.8)1359 (91.6)10,193 (90.9)0.2910.030
Surgical approach<0.0010.039
 Posterior8569 (88.1)1299 (87.6)9868 (88.0)
 Anterior378 (3.9)18 (1.2)396 (3.5)
 Lateral135 (1.4)105 (7.1)240 (2.1)
 2-stage648 (6.7)61 (4.1)709 (6.3)
Predominant symptom<0.0010.193
 Pain8593 (88.3)1392 (93.9)9985 (89.0)
 Weakness320 (3.3)28 (1.9)348 (3.1)
 Numbness/tingling817 (8.4)63 (4.2)880 (7.8)
Predominant location of pain0.3930.035
 Back2796 (28.8)401 (27.0)3197 (28.5)
 Leg2262 (23.3)351 (23.7)2613 (23.3)
 Back & leg4667 (48.0)731 (49.3)5398 (48.2)
Symptom duration, mos0.0110.076
 <3428 (4.4)50 (3.4)478 (4.3)
 >39216 (94.7)1410 (95.1)10,626 (94.8)
 Unknown86 (0.9)23 (1.6)109 (1.0)
Depression2430 (25.0)306 (20.6)2736 (24.4)<0.0010.104
Anxiety1911 (19.6)199 (13.4)2110 (18.8)<0.0010.168

Values are shown as number (percent) or mean (SD) unless indicated otherwise. Boldface type indicates statistical significance (p < 0.05).

Predictors of MIS and Open Surgery

Multivariable logistic regression was performed to identify characteristics independently associated with the decision to perform either MIS or open lumbar fusion. Factors associated with the decision to perform open surgery were revision surgery, American Society of Anesthesiologists Physical Status Classification System (ASA) grades III and IV versus grades I and II, partially dependent ambulatory status, high school education or less, adjacent-segment disease, mechanical disc collapse, pseudarthrosis, ≥ 2 levels treated, 2-stage surgery, weakness/numbness/tingling as the predominant symptoms versus pain, and anxiety. Private insurance status was independently associated with MIS compared with open fusion (Fig. 1).

FIG. 1.
FIG. 1.

Multivariable analysis of characteristics independently associated with patients who underwent MIS. VA = Veterans Affairs. Figure is available in color online only.

Optimal Matching and Key PROs

Patients who underwent MIS for lumbar spinal fusion (n = 1483) were matched in a 1:2 fashion with those who underwent open lumbar spinal fusion (n = 2966) in terms of 33 variables; the baseline characteristics of the matched patients are shown in Table 2. After matching, the mean ± SD age was 61.6 ± 12.0 years, BMI was 30.8 ± 6.3 kg/m2, and 54.3% of patients were female. All SMDs were less than 0.1, indicating sufficient matching between the two groups.

TABLE 2.

Baseline characteristics of the patients who underwent MIS and open lumbar fusion after matching

CharacteristicOpen (n = 2966)MIS (n = 1483)Total (n = 4449)p ValueSMD
Age, yrs61.557 (12.182)61.652 (11.627)61.588 (11.999)0.8030.008
BMI, kg/m231.001 (6.323)30.541 (6.225)30.848 (6.294)0.0220.073
Smoker324 (10.9)162 (10.9)486 (10.9)>0.99<0.001
Female sex1614 (54.4)801 (54.0)2415 (54.3)0.7980.008
Revision329 (11.1)180 (12.1)509 (11.4)0.3020.033
Osteoporosis190 (6.4)84 (5.7)274 (6.2)0.3320.031
Diabetes mellitus530 (17.9)256 (17.3)786 (17.7)0.6170.016
ASA grade0.4490.019
 I94 (3.2)52 (3.5)146 (3.3)
 II1654 (55.8)798 (53.8)2452 (55.1)
 III1195 (40.3)625 (42.1)1820 (40.9)
 IV 23 (0.8)8 (0.5)31 (0.7)
Ambulation0.1370.054
 Independent2597 (87.6)1327 (89.5)3924 (88.2)
 Partially dependent353 (11.9)147 (9.9)500 (11.2)
 Totally dependent16 (0.5)9 (0.6)25 (0.6)
Education level0.0590.031
 High school or less1380 (46.5)652 (44.0)2032 (45.7)
 Graduate level1095 (36.9)608 (41.0)1703 (38.3)
 Postgraduate level431 (14.5)199 (13.4)630 (14.2)
 Not available60 (2.0)24 (1.6)84 (1.9)
Symptomatic disc herniation618 (20.8)287 (19.4)905 (20.3)0.2470.037
Grade I spondylolisthesis1750 (59.0)875 (59.0)2625 (59.0)>0.99<0.001
Lumbar stenosis1563 (52.7)813 (54.8)2376 (53.4)0.1810.043
Adjacent-segment disease157 (5.3)89 (6.0)246 (5.5)0.3300.031
Single-level symptomatic mechanical disc collapse19 (0.6)12 (0.8)31 (0.7)0.5240.020
Pseudarthrosis8 (0.3)6 (0.4)14 (0.3)0.4490.023
ODI at baseline44.651 (15.390)45.061 (15.771)44.788 (15.517)0.4070.026
VAS
 Leg pain6.505 (2.804)6.567 (2.784)6.526 (2.797)0.4860.022
 Back pain6.698 (2.572)6.764 (2.604)6.720 (2.582)0.4190.026
Race
 African American229 (7.7)112 (7.6)341 (7.7)0.8420.006
 White2620 (88.3)1317 (88.8)3937 (88.5)0.6420.015
 Other98 (3.3)34 (2.3)132 (3.0)0.0610.061
Insurance0.0010.023
 Medicare1229 (41.4)617 (41.6)1846 (41.5)
 Medicaid108 (3.6)33 (2.2)141 (3.2)
 Private1485 (50.1)791 (53.3)2276 (51.2)
 Uninsured20 (0.7)4 (0.3)24 (0.5)
 Veterans Affairs/government124 (4.2)38 (2.6)162 (3.6)
Workers’ compensation56 (1.9)23 (1.6)79 (1.8)0.4220.026
Liability claim76 (2.6)43 (2.9)119 (2.7)0.5110.021
No. of levels fused1.233 (0.493)1.220 (0.471)1.229 (0.486)0.3830.028
Levels fused0.4220.028
 12369 (79.9)1194 (80.5)3563 (80.1)
 2502 (16.9)252 (17.0)754 (16.9)
 395 (3.2)37 (2.5)132 (3.0)
Interbody graft2684 (90.5)1359 (91.6)4043 (90.9)0.2110.040
Surgical approach<0.0010.064
 Posterior2583 (87.1)1299 (87.6)3882 (87.3)
 Anterior165 (5.6)18 (1.2)183 (4.1)
 Lateral93 (3.1)105 (7.1)198 (4.5)
 2-stage surgery125 (4.2)61 (4.1)186 (4.2)
Predominant symptom0.3340.021
 Pain2760 (93.1)1392 (93.9)4152 (93.3)
 Weakness77 (2.6)28 (1.9)105 (2.4)
 Numbness/tingling129 (4.3)63 (4.2)192 (4.3)
Predominant location of pain0.8700.017
 Back783 (26.4)401 (27.0)1184 (26.6)
 Leg698 (23.5)351 (23.7)1049 (23.6)
 Back & leg1485 (50.1)731 (49.3)2216 (49.8)
Symptom duration, mos0.8140.014
 <3102 (3.4)50 (3.4)152 (3.4)
 >32825 (95.2)1410 (95.1)4235 (95.2)
 Unknown39 (1.3)23 (1.6)62 (1.4)
Depression631 (21.3)306 (20.6)937 (21.1)0.6210.016
Anxiety348 (11.7)199 (13.4)547 (12.3)0.1060.051

Values are shown as number (percent) or mean (SD) unless indicated otherwise. Boldface type indicates statistical significance (p < 0.05).

Overall, a greater proportion of patients who underwent MIS for lumbar fusion had high satisfaction (NASS score 1–2) at 3 months postoperatively than those who underwent open lumbar fusion (92.9% vs 90.3%, p = 0.004; number needed to treat [NNT] 39). A greater proportion of patients in the MIS group also had maximal satisfaction (NASS score 1) at 3 months than the open lumbar fusion group (74.0% vs 69.8%, p = 0.003; NNT 24). The mean ± SD decrease from baseline ODI at 3 months was significantly greater in the MIS group (21.1 ± 18.4 vs 19.5 ± 18.3, p = 0.006) (Fig. 2). At 12 months postoperatively, overall satisfaction was comparable between MIS and open lumbar fusion patients (86.6% vs 86.1% of patients, p = 0.666); however, decrease from baseline ODI remained significantly greater in the MIS group (24.6 ± 19.2 vs 22.3 ± 19.3, p < 0.0001) (Fig. 2).

FIG. 2.
FIG. 2.

Timelines of changes in PROs of the MIS and open groups. Changes in ODI (A), leg pain VAS (B), and back pain VAS (C) are shown. ns = not significant. **p < 0.01; ****p < 0.0001. Figure is available in color online only.

Overall Satisfaction at 3 Months

Demographic Characteristics

When examining age (< 58, 58–67, and ≥ 68 years), we determined that patients < 58 years had higher odds of achieving satisfaction with MIS than patients who underwent open lumbar fusion (OR 1.82, 95% CI 1.23–2.68, p = 0.003), although all age groups trended toward higher satisfaction with MIS (Fig. 3). Patients with BMI ≥ 30 had high odds of achieving satisfaction with MIS lumbar fusion (OR 1.38, 95% CI 1.01–1.91, p = 0.044). This was also true for those with BMI < 30, although this finding was not significant (OR 1.4, 95% CI 1.00–1.97, p = 0.053). Male sex (OR 1.65, 95% CI 1.17–2.33, p = 0.005) and White race (OR 1.45, 95% CI 1.11–1.88, p = 0.006) were also associated with higher odds of satisfaction after MIS lumbar fusion.

FIG. 3.
FIG. 3.

Comparison of ORs of satisfaction with surgery (NASS score 1 or 2) between patients who underwent MIS and open approaches according to different patient characteristics. Afr. = African; ASD = adjacent-segment disease; DH = disc herniation; Pred. = predominant; Spond. = spondylolisthesis; SSMDC = single-level symptomatic mechanical disc collapse.

When examining age (< 58, 58–67, and ≥ 68 years), we determined that patients < 58 years had higher odds of achieving satisfaction with MIS than patients who underwent open lumbar fusion (OR 1.82, 95% CI 1.23–2.68, p = 0.003), although all age groups trended toward higher satisfaction with MIS (Fig. 3). Patients with BMI ≥ 30 had high odds of achieving satisfaction with MIS lumbar fusion (OR 1.38, 95% CI 1.01–1.91, p = 0.044). This was also true for those with BMI < 30, although this finding was not significant (OR 1.4, 95% CI 1.00–1.97, p = 0.053). Male sex (OR 1.65, 95% CI 1.17–2.33, p = 0.005) and White race (OR 1.45, 95% CI 1.11–1.88, p = 0.006) were also associated with higher odds of satisfaction after MIS lumbar fusion.

Anatomical and Surgical Factors

The absence of grade I spondylolisthesis (OR 1.54, 95% CI 1.09–2.18, p = 0.014) and the presence of lumbar stenosis (OR 1.59, 95% CI 1.17–2.17, p = 0.003) were associated with higher odds of satisfaction after MIS lumbar fusion. Single-level MIS fusion was also associated with higher overall odds of satisfaction (OR 1.54, 95% CI 1.18–2.02, p = 0.002) than multilevel MIS fusion. Primary surgery was associated with higher odds of overall satisfaction with MIS fusion (OR 1.44, 95% CI 1.12–1.86, p = 0.005) than revision surgery (OR 1.27, 95% CI 0.72–2.22, p = 0.408).

Twelve-Month Follow-Up and Other Endpoints

Oswestry Disability Index

Baseline ODI scores were similar between groups (45.1 for the MIS group vs 44.6 for the open surgery group, p = 0.407); however, patients who underwent MIS lumbar fusion had lower ODI scores at 3 months (p = 0.036) and 12 months (p = 0.001) postoperatively than those who underwent open lumbar fusion (Table 3). The decreases from baseline ODI at 3 months (21.1 ± 18.4 vs 19.5 ± 18.3, p = 0.006) and 12 months (24.6 ± 19.2 vs 22.3 ± 19.3; p < 0.0001) were also greater in the MIS group (Fig. 2). The proportion of patients who achieved optimal ODI (< 20) was also significantly greater in the MIS group than the open group at 3 months (47.6% vs 43.8%, p = 0.016; NNT 26) and 12 months (55.2% vs 51.6%, p = 0.026; NNT 32).

TABLE 3.

Postoperative patient satisfaction, change in ODI, and clinical outcomes after MIS and open lumbar fusion

CharacteristicOpen (n = 2966)MIS (n = 1483)Total (n = 4449)p ValueNNT (95% CI)
Primary endpoints
 Satisfaction (NASS score 1–2) at 3 mos2677 (90.3)1377 (92.9)4054 (91.1)0.00439 (23–110)
 Max satisfaction (NASS score 1) at 3 mos2069 (69.8)1097 (74.0)3166 (71.2)0.00324 (14–70)
ODI at baseline44.651 (15.390)45.061 (15.771)44.788 (15.517)0.407
 ODI at 3 mos25.145 (18.155)23.941 (17.792)24.743 (18.042)0.036
 Optimal ODI (<20) at 3 mos1299 (43.8)706 (47.6)2005 (45.1)0.01626 (14–142)
 Decrease from baseline ODI at 3 mos19.506 (18.265)21.120 (18.393)20.044 (18.321)0.006
 MCID (30%) in ODI at 3 mos1934 (65.2)1011 (68.2)2945 (66.2)0.04933 (17–3287)
Secondary endpoints
 Satisfaction (NASS score 1–2) at 12 mos2554 (86.1)1284 (86.6)3838 (86.3)0.666
 Max satisfaction (NASS score 1) at 12 mos1951 (65.8)1004 (67.7)2955 (66.4)0.201
 ODI at 12 mos22.364 (19.226)20.377 (18.695)21.702 (19.071)0.001
 Optimal ODI (<20) at 12 mos1520 (51.6)812 (55.2)2332 (52.8)0.02632 (17–259)
 Decrease from baseline ODI at 12 mos22.289 (19.338)24.639 (19.176)23.072 (19.314)<0.001
 MCID (30%) in ODI at 12 mos2043 (69.4)1098 (74.6)3141 (71.1)<0.00119 (13–41)
 LOS, days2.843 (1.663)2.457 (2.809)2.714 (2.122)<0.001
 Length of surgery, mins172.32 (72.946)176.84 (83.355)173.84 (76.613)0.065
 Incidental durotomy28 (1.4)14 (1.3)42 (1.4)0.677
 Readmission w/in 30 days98 (3.3)48 (3.2)146 (3.3)0.911
 Revision surgery w/in 3 mos76 (2.6)41 (2.8)117 (2.6)0.704
 Revision surgery w/in 12 mos154 (5.6)60 (4.1)214 (5.1)0.03266 (35–511)

Values are shown as number (percent) or mean (SD) unless indicated otherwise. Boldface type indicates statistical significance (p < 0.05).

Leg and Back Pain

Patients who underwent MIS fusion had a greater postoperative decrease than the open fusion group in both back and leg pain at 3 months (decrease in back pain VAS 3.89 vs 3.55, p < 0.001; decrease in leg pain VAS 4.49 vs 4.19, p = 0.008) and 12 months (decrease in back pain 3.96 vs 3.60, p < 0.001; decrease in leg pain 4.45 vs 4.10, p = 0.002) (Fig. 2). A greater proportion of MIS patients also achieved minimal back pain (≤ 2 on VAS) at both 3 months (51.2% vs 46.7%, p = 0.005) and 12 months (53.7% vs 49.7%, p = 0.013) and minimal leg pain at 12 months (66.8% vs 62.5%, p = 0.006) (Table 4).

TABLE 4.

Postoperative changes in leg and back pain (secondary endpoints) after MIS and open lumbar fusion

CharacteristicOpen (n = 2966)MIS (n = 1483)Total (n = 4449)p ValueNNT (95% CI)
VAS for leg pain at baseline6.505 (2.804)6.567 (2.784)6.526 (2.797)0.486
VAS for leg pain at 3 mos2.312 (2.829)2.072 (2.608)2.232 (2.760)0.006
Optimal leg pain (≤2 in VAS) at 3 mos1885 (63.6)982 (66.3)2867 (64.5)0.075
Decrease in leg pain at 3 mos4.192 (3.626)4.494 (3.493)4.293 (3.584)0.008
MCID (30%) in leg pain at 3 mos2297 (77.5)1200 (81.0)3497 (78.7)0.00728 (17–97)
Leg pain at 12 mos2.406 (2.924)2.100 (2.741)2.304 (2.868)<0.001
Optimal leg pain (≤2 in VAS) at 12 mos1844 (62.5)984 (66.8)2828 (63.9)0.00624 (14–80)
Decrease in leg pain at 12 mos4.103 (3.667)4.453 (3.509)4.220 (3.618)0.002
MCID (30%) in leg pain at 12 mos2221 (75.3)1181 (80.1)3402 (76.9)<0.00121 (14–45)
VAS for back pain at baseline6.698 (2.572)6.764 (2.604)6.720 (2.582)0.419
VAS for back pain at 3 mos3.145 (2.600)2.874 (2.447)3.055 (2.553)<0.001
Optimal back pain (≤2 in VAS) at 3 mos1386 (46.7)758 (51.2)2144 (48.2)0.00523 (13–76)
Decrease in back pain at 3 mos3.553 (3.049)3.889 (3.089)3.665 (3.066)<0.001
MCID (30%) in back pain at 3 mos2157 (72.7)1132 (76.4)3289 (74.0)0.00827 (17–100)
Back pain at 12 mos3.134 (2.839)2.805 (2.663)3.024 (2.786)<0.001
Optimal back pain (≤2 in VAS) at 12 mos1466 (49.7)791 (53.7)2257 (51.0)0.01325 (14–120)
Decrease in back pain at 12 mos3.569 (3.188)3.955 (3.197)3.698 (3.196)<0.001
MCID (30%) in back pain at 12 mos2096 (71.1)1101 (74.7)3197 (72.3)0.01128 (16–116)

Values are shown as number (percent) or mean (SD) unless indicated otherwise. Boldface type indicates statistical significance (p < 0.05).

Minimal Clinically Important Difference

An MCID of 30% in ODI was achieved in a greater proportion of MIS patients than open fusion patients at 3 months (68.2% vs 65.2% of patients, p = 0.049) and 12 months (74.6% vs 69.4%, p < 0.001). MCID for back and leg pain was also achieved by a higher proportion of MIS patients than open fusion patients at both 3 months (76.4% vs 72.7% for back pain, p = 0.008; 81.0% vs 77.5% for leg pain, p = 0.007) and 12 months (74.7% vs 71.1% for back pain, p = 0.011; 80.1% vs 75.3% for leg pain, p < 0.001) (Tables 3 and 4).

Other Clinical Outcomes

LOS was significantly shorter for the MIS lumbar fusion group than for the open fusion group (2.46 vs 2.84 days, p < 0.001). Revision surgery within 12 months of initial surgery was performed on a greater proportion of patients in the open surgery group (5.6% vs 4.1%, p = 0.032; NNT 66); however, the proportions of patients who underwent revision surgery within 3 months were similar between groups (2.6% vs 2.8%, p = 0.704) (Table 3). Length of surgery, incidental durotomy rates, and readmission within 30 days were not significantly different between groups.

Subgroup Analysis

Clinical outcomes were examined in subgroups determined on the basis of indication for surgery. Key PROs for the various surgical indications are displayed in Table 5.

TABLE 5.

Subgroup analysis of key PROs according to indication for surgery

IndicationOpenMISp Value
Lumbar stenosis (n = 2376)1563813
 Satisfaction (NASS score 1–2) at 3 mos1390 (88.9)754 (92.7)0.003
 Satisfaction (NASS score 1–2) at 12 mos1338 (85.6)710 (87.3)0.247
 Decrease from baseline ODI at 3 mos18.773 (17.939)20.044 (18.230)0.103
 Decrease from baseline ODI at 12 mos21.341 (19.324)23.824 (18.727)0.003
 Decrease in leg pain at 12 mos4.039 (3.616)4.522 (3.502)0.002
 Decrease in back pain at 12 mos3.516 (3.204)4.037 (3.218)<0.001
Grade I spondylolisthesis (n = 2625)1750875
 Satisfaction (NASS score 1–2) at 3 mos1600 (91.4)816 (93.3)0.103
 Satisfaction (NASS score 1–2) at 12 mos1540 (88.0)770 (88.0)>0.99
 Decrease from baseline ODI at 3 mos19.800 (18.179)20.912 (18.173)0.140
 Decrease from baseline ODI at 12 mos22.599 (19.044)25.124 (18.855)0.001
 Decrease in leg pain at 12 mos4.294 (3.686)4.525 (3.557)0.128
 Decrease in back pain at 12 mos3.734 (3.266)4.165 (3.200)0.001
Disc herniation (n = 905)618287
 Satisfaction (NASS score 1–2) at 3 mos549 (88.8)262 (91.3)0.260
 Satisfaction (NASS score 1–2) at 12 mos522 (84.5)250 (87.1)0.296
 Decrease from baseline ODI at 3 mos20.116 (18.663)21.772 (19.115)0.218
 Decrease from baseline ODI at 12 mos23.175 (20.015)25.580 (19.650)0.093
 Decrease in leg pain at 12 mos3.859 (3.643)4.369 (3.345)0.046
 Decrease in back pain at 12 mos3.293 (3.118)3.692 (3.077)0.074
Adjacent-segment disease (n = 246)15789
 Satisfaction (NASS score 1–2) at 3 mos134 (85.4)85 (95.5)0.014
 Satisfaction (NASS score 1–2) at 12 mos119 (75.8)78 (87.6)0.025
 Decrease from baseline ODI at 3 mos16.937 (16.464)21.930 (17.776)0.027
 Decrease from baseline ODI at 12 mos17.614 (18.973)20.985 (18.730)0.183
 Decrease in leg pain at 12 mos4.090 (3.963)3.977 (3.038)0.818
 Decrease in back pain at 12 mos2.872 (3.311)3.368 (2.910)0.244

Values are shown as number, number (percent), or mean (SD) unless indicated otherwise. Boldface type indicates statistical significance (p < 0.05).

Lumbar Stenosis

Patients with lumbar stenosis achieved greater satisfaction at 3 months after MIS than patients who underwent open surgery (p = 0.003) and also demonstrated a greater decrease in ODI at 12 months (p = 0.003). Decreases in leg and back pain (both overall and MCID) were also significantly greater in patients who underwent MIS compared with those who underwent open surgery.

Grade I Spondylolisthesis

Patients with grade I spondylolisthesis demonstrated a greater decrease from baseline ODI at 12 months after MIS, but these patients did not demonstrate significant differences in satisfaction at 3 or 12 months or change in ODI at 3 months. Decreases in leg and back pain at 3 and 12 months were greater with MIS compared with open fusion.

Disc Herniation

There were no significant differences in the key endpoints between MIS and open fusion; however, patients who underwent MIS had significantly greater improvements in leg pain (both decrease and overall MCID at 3 and 12 months) and back pain at 3 months.

Adjacent-Segment Disease

Satisfaction was significantly higher at 3 and 12 months, and decrease in ODI was greater at 3 months, in patients who underwent MIS. Additionally, MCID for leg pain was greater at 3 months in patients who underwent MIS.

Discussion

The indications for the treatment of degenerative lumbar pathology with MIS technologies continue to expand as these procedures become more and more popular. MIS approaches involve tissue-sparing techniques with smaller incisions and have been associated with diminished blood loss, operative times, and hospital LOS.14–18 Prior studies have also shown that MIS approaches have equivalent surgical complications to open procedures, but with a lower rate of medical complications.19

To date, this is the largest multicenter database study to examine patient outcomes after MIS versus those after open lumbar spinal fusion by utilizing the lumbar spine surgery module of the QOD. After logistic regression analysis and optimal 1:2 matching for 33 baseline variables, ORs for overall satisfaction with MIS lumbar fusion versus those for open lumbar fusion were calculated across several patient characteristics in an effort to determine the patient populations most likely to benefit from MIS for lumbar fusion. Overall, MIS for lumbar fusion was found to be superior to open lumbar fusion in terms of satisfaction at 3 months postoperatively, but no such difference was observed at 12 months.

Several prior single-institution and multicenter studies have been performed to compare MIS and open treatment of degenerative lumbar disease; however, the majority have demonstrated similar improvements regardless of approach.16–18 Mummaneni et al. previously utilized the QOD to examine MIS versus open single-level fusion for grade I degenerative lumbar spondylolisthesis, but no significant differences between groups were found; significant improvement in all functional outcomes was noted in both groups.6 Similarly, McGirt et al. examined patients who underwent either MIS or open 1- to 2-level interbody fusion for lumbar stenosis or spondylolisthesis, and they found reduced blood loss and LOS in patients who underwent MIS without associated improvements in 12-month outcomes.4 Examining patients identified in the QOD who underwent only MIS versus open decompression for low-grade spondylolisthesis, Bisson et al. found comparable PROs and clinical outcomes at 2 years.3

Although these prior similar studies have also utilized the QOD, some features differentiate the index study. More specifically, in this study, we utilized a significantly larger and demographically heterogeneous volume of patients, which adds significant power to the comparative analysis and enhances the generalizability of the results. Additionally, contrary to previous studies that used multivariable analysis to adjust for confounders, in this study we performed optimal matching to adjust for confounders. This method allowed us to adjust for confounders directly, potentially providing significant mitigation of selection bias.10

This is one of the first studies to demonstrate a statistically significant higher odds of short-term (3-month) satisfaction with MIS lumbar fusion compared with open lumbar fusion, with similar overall satisfaction at 12 months. In addition, MIS lumbar fusion was associated with greater decreases in the ODI and back and leg pain, a greater percentage of patients who achieved optimal ODI and optimal back pain, and a greater percentage of patients who achieved MCID of 30% for back pain, leg pain, and ODI at both 3 and 12 months postoperatively. Despite the robust statistical significance of these findings, their clinical impact is limited because of the small absolute values of these differences, potentially disputing the clinical significance of these findings. However, given the high similarity between the two groups, differences in outcomes should be predominantly attributed to the surgical technique, and such differences are not expected to be vast.

There is a learning curve for the optimal implementation of all emerging technologies, and it may be that MIS technology is reaching its “plateau of productivity” with the increasing realization of its benefits.20 Assessment of a learning curve was not feasible in the index study because the majority of the MIS operations took place in a relatively short period (2015–2018). Recent improvements in MIS technology, in combination with its more widespread use and increasing familiarity among surgeons, may explain the present novel findings.

This study also demonstrated a specific benefit of MIS fusion for several patient subgroups, including those with age ≤ 55 years, men, White patients, those undergoing first-time surgery or single-level surgery, those with lumbar stenosis, and those without spondylolisthesis. Additionally, on the basis of 3-month satisfaction, subgroup analysis demonstrated that patients who underwent surgery for lumbar stenosis may be more likely to benefit from an MIS approach than those with spondylolisthesis or a disc herniation. However, it should be noted that all patients had significantly greater improvements in satisfaction and ODI with an MIS approach, regardless of surgical indication. It is likely that the numbers of patients in some subgroups were not large enough to reveal significant differences between surgical approaches. Increased understanding of the preoperative factors that may influence optimal patient outcomes may aid physicians in counseling patients on the various surgical approaches for a specific pathology and choosing the optimal surgery for each patient.

Interestingly, private insurance status and education level were found to be independent predictors of the decision to perform MIS lumbar fusion on multivariate analysis, highlighting potential socioeconomic drivers for the use of MIS or open surgery.

Prior studies have shown that depression and anxiety impact short-term but not long-term PROs after lumbar surgery for spondylolisthesis.21 Our study demonstrated improved 3-month satisfaction with MIS in both those with and those without anxiety and depression, although significance was not demonstrated likely because of inadequate sample size.

Interestingly, obesity (BMI ≥ 30) and higher ASA grades III and IV were significantly associated with improved satisfaction with MIS. Prior studies showed that higher BMI was associated with longer operative times, greater blood loss, and higher risks of surgical site infection and nerve injury, in addition to poorer PROs and satisfaction at 1 to 2 years of follow-up.22–27 However, the tissue-sparing effects of MIS fusion may be more pronounced in obese patients, resulting in the greater improvement in satisfaction after MIS observed in this particular subgroup. Indeed, one prior systematic review demonstrated that although obese patients had greater surgical blood loss, longer operative times, and higher complication/reoperation rates, these differences were not significant in the subgroup of obese patients who underwent MIS.23 Our study further corroborates these data.

Similar to most prior studies, the present study found that patients who underwent MIS fusion had an LOS that was approximately a half-day shorter than patients who underwent open surgery, with similar surgery times for both groups. Additionally, the durotomy rates, readmission rates, and rates of revision surgery within 3 months were not significantly different between groups, but the rate of revision surgery within 12 months was higher in the open group. Prior studies have demonstrated mixed results for durotomy rates between MIS and open approaches.28–30

Limitations

The limitations and weaknesses inherent to the current study have implications for its interpretation. First, this was an observational study, and patients were not randomly assigned to treatment groups. Although we employed optimal matching to reduce potential selection bias and the effects of confounders, it is possible that variables that were not collected and adjusted for could have affected the results. This is an inherent limitation of observational studies. Despite the final cohorts differing in terms of BMI and insurance status after optimal matching, SMD remained < 0.1 for all categories, thereby indicating optimal matching efficiency. Second, our length of follow-up was 1 year; as such, we can make no statements regarding the long-term comparative durability of MIS versus open technologies.

Additionally, inclusion of only patients with available 3- and 12-month follow-up could be another source of selection bias. Lastly, it is important to recognize that this is not an analysis of a specific MIS versus open approach, but rather an analysis of how surgeons in everyday care use MIS technologies in a variety of MIS approaches and how outcomes may be affected by those applications. The possibility that the surgeons who performed MIS were highly trained in this technique and, therefore, more likely to achieve optimal outcomes should be recognized. This scenario possibly limits the applicability of our conclusions to surgeons with less experience in MIS. Nevertheless, we included a diverse group of practices and patients to obtain a representative sample of patients who underwent elective lumbar spinal surgery, allowing these results to provide valuable evidence from a population health perspective.

Conclusions

In patients who underwent lumbar fusion for degenerative spine disease, MIS lumbar fusion was associated with higher odds of satisfaction and a greater decrease in ODI than open lumbar fusion at 3 months postoperatively. No statistically significant difference was demonstrated between patients who underwent MIS and those who underwent open lumbar fusion in terms of satisfaction at 12 months, but those who underwent MIS possibly maintained greater improvement in ODI. MIS was associated with a slightly greater decrease in back and leg pain, as well as lower 12-month reoperation rates. Additionally, overall durotomy and readmission rates are similar regardless of approach, with a shorter average LOS in patients who underwent MIS lumbar fusion. Independent factors associated with increased satisfaction with MIS lumbar fusion, compared with open surgery, included age < 58 years, first-time surgery, single-level surgery, lumbar stenosis, and absence of spondylolisthesis.

Disclosures

Dr. Mummaneni is a consultant for DePuy Synthes, Globus, and Stryker; owns stock in Spinicity/ISD; receives clinical or research support for the study described from NREF; receives non–study-related clinical or research support from AO Spine and ISSG; and receives royalties from DePuy Synthes, Thieme Publishers, and Springer Publishers. Dr. Bisson is a consultant for MiRus, Stryker, and Medtronic; and owns stock in MiRus and nView. Dr. Haid is a consultant for NuVasive; receives royalties from IP Royalties: Globus Medical, Medtronic, and NuVasive; and owns stock in Globus Medical, NuVasive, and Remedy Health Media. Dr. Knightly is chair of the board of directors of NPA. Dr. Devin is a consultant for Stryker and owns stock in 3Spine. Dr. Pennicooke is a consultant for Pacira Biosciences, Inc.

Author Contributions

Conception and design: Mooney, Michalopoulos. Acquisition of data: Mooney. Analysis and interpretation of data: Michalopoulos. Drafting the article: Mooney, Michalopoulos, Zeitouni. Critically revising the article: Bydon, Chan, Mummaneni, Bisson, Haid, Pennicooke. Reviewed submitted version of manuscript: Bydon, Alvi, Chan, Mummaneni, Sherrod, Knightly, Devin, Asher. Statistical analysis: Michalopoulos, Alvi. Study supervision: Bydon, Bisson, Devin, Pennicooke, Asher.

References

  • 1

    Hammad A, Wirries A, Ardeshiri A, Nikiforov O, Geiger F. Open versus minimally invasive TLIF: literature review and meta-analysis. J Orthop Surg Res. 2019;14(1):229.

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 2

    Kulkarni AG, Bohra H, Dhruv A, Sarraf A, Bassi A, Patil VM. Minimal invasive transforaminal lumbar interbody fusion versus open transforaminal lumbar interbody fusion. Indian J Orthop. 2016;50(5):464472.

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 3

    Bisson EF, Mummaneni PV, Virk MS, Knightly J, Alvi MA, Goyal A, et al. Open versus minimally invasive decompression for low-grade spondylolisthesis: analysis from the Quality Outcomes Database. J Neurosurg Spine. 2020;33(3):349359.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 4

    McGirt MJ, Parker SL, Mummaneni P, Knightly J, Pfortmiller D, Foley K, Asher AL. Is the use of minimally invasive fusion technologies associated with improved outcomes after elective interbody lumbar fusion? Analysis of a nationwide prospective patient-reported outcomes registry. Spine J. 2017;17(7):922932.

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 5

    Tschugg A, Hartmann S, Lener S, Rietzler A, Sabrina N, Thomé C. Minimally invasive spine surgery in lumbar spondylodiscitis: a retrospective single-center analysis of 67 cases. Eur Spine J. 2017;26(12):31413146.

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 6

    Mummaneni PV, Bisson EF, Kerezoudis P, Glassman S, Foley K, Slotkin JR, et al. Minimally invasive versus open fusion for Grade I degenerative lumbar spondylolisthesis: analysis of the Quality Outcomes Database. Neurosurg Focus. 2017;43(2):E11.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 7

    Finkelstein JA, Schwartz CE. Patient-reported outcomes in spine surgery: past, current, and future directions. J Neurosurg Spine. 2019;31(2):155164.

  • 8

    Asher AM, Oleisky ER, Pennings JS, Khan I, Sivaganesan A, Devin CJ, et al. Measuring clinically relevant improvement after lumbar spine surgery: is it time for something new? Spine J. 2020;20(6):847856.

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 9

    R Core Team. R: A Language and Environment for Statistical Computing. Accessed October 12, 2021. https://www.R-project.org/

  • 10

    Kallus N, Pennicooke B, Santacatterina M. More robust estimation of average treatment effects using kernel optimal matching in an observational study of spine surgical interventions. Stat Med. 2021;40(10):23052320.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 11

    Stuart EA. Matching methods for causal inference: a review and a look forward. Stat Sci. 2010;25(1):121.

  • 12

    Kimachi M, Onishi A, Tajika A, Kimachi K, Furukawa TA. Systematic differences in effect estimates between observational studies and randomized control trials in meta-analyses in nephrology. Sci Rep. 2021;11(1):6088.

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 13

    Fortin SP, Johnston SS, Schuemie MJ. Applied comparison of large-scale propensity score matching and cardinality matching for causal inference in observational research. BMC Med Res Methodol. 2021;21(1):109.

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 14

    Parker SL, Mendenhall SK, Shau DN, Zuckerman SL, Godil SS, Cheng JS, McGirt MJ. Minimally invasive versus open transforaminal lumbar interbody fusion for degenerative spondylolisthesis: comparative effectiveness and cost-utility analysis. World Neurosurg. 2014;82(1-2):230238.

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 15

    Shunwu F, Xing Z, Fengdong Z, Xiangqian F. Minimally invasive transforaminal lumbar interbody fusion for the treatment of degenerative lumbar diseases. Spine (Phila Pa 1976).2010;35(17):16151620.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 16

    Schizas C, Tzinieris N, Tsiridis E, Kosmopoulos V. Minimally invasive versus open transforaminal lumbar interbody fusion: evaluating initial experience. Int Orthop. 2009;33(6):16831688.

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 17

    Peng CW, Yue WM, Poh SY, Yeo W, Tan SB. Clinical and radiological outcomes of minimally invasive versus open transforaminal lumbar interbody fusion. Spine (Phila Pa 1976).2009;34(13):13851389.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 18

    Dhall SS, Wang MY, Mummaneni PV. Clinical and radiographic comparison of mini-open transforaminal lumbar interbody fusion with open transforaminal lumbar interbody fusion in 42 patients with long-term follow-up. J Neurosurg Spine. 2008;9(6):560565.

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 19

    Goldstein CL, Macwan K, Sundararajan K, Rampersaud YR. Perioperative outcomes and adverse events of minimally invasive versus open posterior lumbar fusion: meta-analysis and systematic review. J Neurosurg Spine. 2016;24(3):416427.

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 20

    Gartner. Gartner Hype Cycle. Accessed October 12, 2021. https://www.gartner.com/en/research/methodologies/gartner-hype-cycle

  • 21

    Kashlan O, Swong K, Alvi MA, Bisson EF, Mummaneni PV, Knightly J, et al. Patients with a depressive and/or anxiety disorder can achieve optimum long term outcomes after surgery for grade 1 spondylolisthesis: analysis from the Quality Outcomes Database (QOD). Clin Neurol Neurosurg. 2020;197:106098.

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 22

    Park C, Garcia AN, Cook C, Shaffrey CI, Gottfried ON. Long-term impact of obesity on patient-reported outcomes and patient satisfaction after lumbar spine surgery: an observational study. J Neurosurg Spine. 2021;34(1):7382.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 23

    Goyal A, Elminawy M, Kerezoudis P, Lu VM, Yolcu Y, Alvi MA, Bydon M. Impact of obesity on outcomes following lumbar spine surgery: a systematic review and meta-analysis. Clin Neurol Neurosurg. 2019;177:2736.

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 24

    Onyekwelu I, Glassman SD, Asher AL, Shaffrey CI, Mummaneni PV, Carreon LY. Impact of obesity on complications and outcomes: a comparison of fusion and nonfusion lumbar spine surgery. J Neurosurg Spine. 2017;26(2):158162.

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 25

    Jackson KL II, Devine JG. The effects of obesity on spine surgery: a systematic review of the literature. Global Spine J. 2016;6(4):394400.

  • 26

    Cao J, Kong L, Meng F, Zhang Y, Shen Y. Impact of obesity on lumbar spinal surgery outcomes. J Clin Neurosci. 2016;28:16.

  • 27

    De la Garza-Ramos R, Bydon M, Abt NB, Sciubba DM, Wolinsky JP, Bydon A, et al. The impact of obesity on short- and long-term outcomes after lumbar fusion. Spine (Phila Pa 1976).2015;40(1):5661.

    • Search Google Scholar
    • Export Citation
  • 28

    Ghobrial GM, Theofanis T, Darden BV, Arnold P, Fehlings MG, Harrop JS. Unintended durotomy in lumbar degenerative spinal surgery: a 10-year systematic review of the literature. Neurosurg Focus. 2015;39(4):E8.

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 29

    Wong AP, Shih P, Smith TR, Slimack NP, Dahdaleh NS, Aoun SG, et al. Comparison of symptomatic cerebral spinal fluid leak between patients undergoing minimally invasive versus open lumbar foraminotomy, discectomy, or laminectomy. World Neurosurg. 2014;81(3-4):634640.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 30

    Evaniew N, Khan M, Drew B, Kwok D, Bhandari M, Ghert M. Minimally invasive versus open surgery for cervical and lumbar discectomy: a systematic review and meta-analysis. CMAJ Open. 2014;2(4):E295E305.

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation

Illustration from Lee et al. (pp 822–829). Copyright Sun Joo Kim. Published with permission.

  • View in gallery

    Multivariable analysis of characteristics independently associated with patients who underwent MIS. VA = Veterans Affairs. Figure is available in color online only.

  • View in gallery

    Timelines of changes in PROs of the MIS and open groups. Changes in ODI (A), leg pain VAS (B), and back pain VAS (C) are shown. ns = not significant. **p < 0.01; ****p < 0.0001. Figure is available in color online only.

  • View in gallery

    Comparison of ORs of satisfaction with surgery (NASS score 1 or 2) between patients who underwent MIS and open approaches according to different patient characteristics. Afr. = African; ASD = adjacent-segment disease; DH = disc herniation; Pred. = predominant; Spond. = spondylolisthesis; SSMDC = single-level symptomatic mechanical disc collapse.

  • 1

    Hammad A, Wirries A, Ardeshiri A, Nikiforov O, Geiger F. Open versus minimally invasive TLIF: literature review and meta-analysis. J Orthop Surg Res. 2019;14(1):229.

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 2

    Kulkarni AG, Bohra H, Dhruv A, Sarraf A, Bassi A, Patil VM. Minimal invasive transforaminal lumbar interbody fusion versus open transforaminal lumbar interbody fusion. Indian J Orthop. 2016;50(5):464472.

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 3

    Bisson EF, Mummaneni PV, Virk MS, Knightly J, Alvi MA, Goyal A, et al. Open versus minimally invasive decompression for low-grade spondylolisthesis: analysis from the Quality Outcomes Database. J Neurosurg Spine. 2020;33(3):349359.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 4

    McGirt MJ, Parker SL, Mummaneni P, Knightly J, Pfortmiller D, Foley K, Asher AL. Is the use of minimally invasive fusion technologies associated with improved outcomes after elective interbody lumbar fusion? Analysis of a nationwide prospective patient-reported outcomes registry. Spine J. 2017;17(7):922932.

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 5

    Tschugg A, Hartmann S, Lener S, Rietzler A, Sabrina N, Thomé C. Minimally invasive spine surgery in lumbar spondylodiscitis: a retrospective single-center analysis of 67 cases. Eur Spine J. 2017;26(12):31413146.

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 6

    Mummaneni PV, Bisson EF, Kerezoudis P, Glassman S, Foley K, Slotkin JR, et al. Minimally invasive versus open fusion for Grade I degenerative lumbar spondylolisthesis: analysis of the Quality Outcomes Database. Neurosurg Focus. 2017;43(2):E11.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 7

    Finkelstein JA, Schwartz CE. Patient-reported outcomes in spine surgery: past, current, and future directions. J Neurosurg Spine. 2019;31(2):155164.

  • 8

    Asher AM, Oleisky ER, Pennings JS, Khan I, Sivaganesan A, Devin CJ, et al. Measuring clinically relevant improvement after lumbar spine surgery: is it time for something new? Spine J. 2020;20(6):847856.

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 9

    R Core Team. R: A Language and Environment for Statistical Computing. Accessed October 12, 2021. https://www.R-project.org/

  • 10

    Kallus N, Pennicooke B, Santacatterina M. More robust estimation of average treatment effects using kernel optimal matching in an observational study of spine surgical interventions. Stat Med. 2021;40(10):23052320.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 11

    Stuart EA. Matching methods for causal inference: a review and a look forward. Stat Sci. 2010;25(1):121.

  • 12

    Kimachi M, Onishi A, Tajika A, Kimachi K, Furukawa TA. Systematic differences in effect estimates between observational studies and randomized control trials in meta-analyses in nephrology. Sci Rep. 2021;11(1):6088.

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 13

    Fortin SP, Johnston SS, Schuemie MJ. Applied comparison of large-scale propensity score matching and cardinality matching for causal inference in observational research. BMC Med Res Methodol. 2021;21(1):109.

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 14

    Parker SL, Mendenhall SK, Shau DN, Zuckerman SL, Godil SS, Cheng JS, McGirt MJ. Minimally invasive versus open transforaminal lumbar interbody fusion for degenerative spondylolisthesis: comparative effectiveness and cost-utility analysis. World Neurosurg. 2014;82(1-2):230238.

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 15

    Shunwu F, Xing Z, Fengdong Z, Xiangqian F. Minimally invasive transforaminal lumbar interbody fusion for the treatment of degenerative lumbar diseases. Spine (Phila Pa 1976).2010;35(17):16151620.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 16

    Schizas C, Tzinieris N, Tsiridis E, Kosmopoulos V. Minimally invasive versus open transforaminal lumbar interbody fusion: evaluating initial experience. Int Orthop. 2009;33(6):16831688.

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 17

    Peng CW, Yue WM, Poh SY, Yeo W, Tan SB. Clinical and radiological outcomes of minimally invasive versus open transforaminal lumbar interbody fusion. Spine (Phila Pa 1976).2009;34(13):13851389.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 18

    Dhall SS, Wang MY, Mummaneni PV. Clinical and radiographic comparison of mini-open transforaminal lumbar interbody fusion with open transforaminal lumbar interbody fusion in 42 patients with long-term follow-up. J Neurosurg Spine. 2008;9(6):560565.

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 19

    Goldstein CL, Macwan K, Sundararajan K, Rampersaud YR. Perioperative outcomes and adverse events of minimally invasive versus open posterior lumbar fusion: meta-analysis and systematic review. J Neurosurg Spine. 2016;24(3):416427.

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 20

    Gartner. Gartner Hype Cycle. Accessed October 12, 2021. https://www.gartner.com/en/research/methodologies/gartner-hype-cycle

  • 21

    Kashlan O, Swong K, Alvi MA, Bisson EF, Mummaneni PV, Knightly J, et al. Patients with a depressive and/or anxiety disorder can achieve optimum long term outcomes after surgery for grade 1 spondylolisthesis: analysis from the Quality Outcomes Database (QOD). Clin Neurol Neurosurg. 2020;197:106098.

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 22

    Park C, Garcia AN, Cook C, Shaffrey CI, Gottfried ON. Long-term impact of obesity on patient-reported outcomes and patient satisfaction after lumbar spine surgery: an observational study. J Neurosurg Spine. 2021;34(1):7382.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 23

    Goyal A, Elminawy M, Kerezoudis P, Lu VM, Yolcu Y, Alvi MA, Bydon M. Impact of obesity on outcomes following lumbar spine surgery: a systematic review and meta-analysis. Clin Neurol Neurosurg. 2019;177:2736.

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 24

    Onyekwelu I, Glassman SD, Asher AL, Shaffrey CI, Mummaneni PV, Carreon LY. Impact of obesity on complications and outcomes: a comparison of fusion and nonfusion lumbar spine surgery. J Neurosurg Spine. 2017;26(2):158162.

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 25

    Jackson KL II, Devine JG. The effects of obesity on spine surgery: a systematic review of the literature. Global Spine J. 2016;6(4):394400.

  • 26

    Cao J, Kong L, Meng F, Zhang Y, Shen Y. Impact of obesity on lumbar spinal surgery outcomes. J Clin Neurosci. 2016;28:16.

  • 27

    De la Garza-Ramos R, Bydon M, Abt NB, Sciubba DM, Wolinsky JP, Bydon A, et al. The impact of obesity on short- and long-term outcomes after lumbar fusion. Spine (Phila Pa 1976).2015;40(1):5661.

    • Search Google Scholar
    • Export Citation
  • 28

    Ghobrial GM, Theofanis T, Darden BV, Arnold P, Fehlings MG, Harrop JS. Unintended durotomy in lumbar degenerative spinal surgery: a 10-year systematic review of the literature. Neurosurg Focus. 2015;39(4):E8.

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 29

    Wong AP, Shih P, Smith TR, Slimack NP, Dahdaleh NS, Aoun SG, et al. Comparison of symptomatic cerebral spinal fluid leak between patients undergoing minimally invasive versus open lumbar foraminotomy, discectomy, or laminectomy. World Neurosurg. 2014;81(3-4):634640.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 30

    Evaniew N, Khan M, Drew B, Kwok D, Bhandari M, Ghert M. Minimally invasive versus open surgery for cervical and lumbar discectomy: a systematic review and meta-analysis. CMAJ Open. 2014;2(4):E295E305.

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation

Metrics

All Time Past Year Past 30 Days
Abstract Views 0 0 0
Full Text Views 625 625 625
PDF Downloads 536 536 536
EPUB Downloads 0 0 0