Decision making for patients with spontaneous intracerebral hemorrhage (ICH) poses several challenges. Outcomes in this patient population are generally poor, prognostication is often uncertain, and treatment strategies offer limited benefits. Studies demonstrate variability in the type and intensity of treatment offered, which is attributed to clinical uncertainty and habits of training. Research has focused on new techniques and more stringent evidence-based selection criteria to improve outcomes and produce consensus around treatment strategies for patients with ICH. Such focus, however, offers little description of how ICH treatment decisions are made and how such decisions reflect patient preferences regarding medical care. A growing body of literature suggests that the process of decision making in ICH is laden with bias, value assumptions, and subjective impressions. Factors such as geography, cognitive biases, patient perceptions, and physician characteristics can all shape decision making and the selection of treatment. Such factors often serve as a barrier to providing patient-centered medical care. In this article, the authors review how surgical decision making for patients with ICH is shaped by these decisional factors and suggest future research pathways to study decision making in ICH. Such research efforts are important for establishing quality guidelines and pay-for-performance measures that reflect the preferences of individual patients and the contextual nature of medical decision making.
Michael L. Kelly, Daniel P. Sulmasy and Robert J. Weil
Michael P. Kelly, Paul A. Anderson, Rick C. Sasso and K. Daniel Riew
The aim of this study is to evaluate the relationship between preoperative opioid strength and outcomes of anterior cervical decompressive surgery.
A retrospective cohort of 1004 patients enrolled in 1 of 2 investigational device exemption studies comparing cervical total disc arthroplasty (TDA) and anterior cervical discectomy and fusion (ACDF) for single-level cervical disease causing radiculopathy or myelopathy was selected. At a preoperative visit, opioid use data, Neck Disability Index (NDI) scores, 36-Item Short-Form Health Survey (SF-36) scores, and numeric rating scale scores for neck and arm pain were collected. Patients were divided into strong (oxycodone/morphine/meperidine), weak (codeine/propoxyphene/hydrocodone), and opioid-naïve groups. Preoperative and postoperative (24 months) outcomes scores were compared within and between groups using the paired t-test and ANCOVA, respectively.
Patients were categorized as follows: 226 strong, 762 weak, and 16 opioid naïve. The strong and weak groups were similar with respect to age, sex, race, marital status, education level, Worker’s Compensation status, litigation status, and alcohol use. At 24-month follow-up, no differences in change in arm or neck pain scores (arm: strong −52.3, weak −50.6, naïve −54.0, p = 0.244; neck: strong −52.7, weak −50.8, naïve −44.6, p = 0.355); NDI scores (strong −36.0, weak −33.3, naïve −32.3, p = 0.181); or SF-36 Physical Component Summary scores (strong: 14.1, weak 13.3, naïve 21.7, p = 0.317) were present. Using a 15-point improvement in NDI to determine success, the authors found no between-groups difference in success rates (strong 80.6%, weak 82.7%, naïve 73.3%, p = 0.134). No difference existed between treatment arms (TDA vs ACDF) for any outcome at any time point.
Preoperative opioid strength did not adversely affect outcomes in this analysis. Careful patient selection can yield good results in this patient population.
Max S. Riley, Keith H. Bridwell, Lawrence G. Lenke, Jonathan Dalton and Michael P. Kelly
Significant health-related quality of life (HRQOL) benefits have been observed for patients undergoing primary and revision adult spinal deformity (ASD) surgery. The purpose of this study was to report changes in HRQOL measures in a consecutive series of patients undergoing complex spinal reconstructive surgery, using Scoli-RISK-1 (SR-1) inclusion criteria.
This was a single-center, retrospective cohort study. The SR-1 inclusion criteria were used to define patients with complex ASD treated between June 1, 2009, and June 1, 2011. Standard preoperative and perioperative data were collected, including the Scoliosis Research Society (SRS)–22r instrument. The HRQOL changes were evaluated at a minimum 2-year follow-up. Standardized forms were used to collect surgery-related complications data for all patients. Complications were defined as minor, transient major, or permanent major. Patients who achieved a minimum 2-year follow-up were included in the analysis.
Eighty-four patients meeting SR-1 criteria were identified. Baseline demographic and surgical data were available for 74/84 (88%) patients. Forty-seven of 74 (64%) patients met the additional HRQOL criteria with a minimum 2-year follow-up (mean follow-up 3.4 years, range 2–6.5 years). Twenty-one percent of patients underwent posterior fusion only, 40% of patients had a posterior column osteotomy, and 38% had a 3-column osteotomy. Seventy-five percent of patients underwent a revision procedure. Significant improvements were observed in all SRS-22r domains: Pain: +0.8 (p < 0.001); Self-Image: +1.4 (p < 0.001); Function: +0.46 (p < 0.001); Satisfaction: +1.6 (p < 0.001); and Mental Health: +0.28 (p = 0.04). With the exception of Mental Health, more than 50% of patients achieved a minimum clinically important difference (MCID) in SRS-22r domain scores (Mental Health: 20/47, 42.6%). A total of 65 complications occurred in 31 patients. This includes 29.8% (14/47) of patients who suffered a major complication and 17% (8/47) who suffered a postoperative neurological deficit, most commonly at the root level (10.6%, 5/47). Of the 8 patients who suffered a neurological deficit, 1 (13%) was able to achieve MCID in the SRS Function domain.
The majority of patients experienced clinically relevant improvement in SRS-22r HRQOL scores after complex ASD surgery. The greatest improvements were seen in the SRS Pain and SRS Self-Image domains. Although 30% of patients suffered a major or permanent complication, benefits from surgery were still attained. Patients sustaining a neurological deficit or major complication were unlikely to achieve HRQOL improvements meeting or exceeding MCID for the SRS Function domain.
Eitan M. Kohan, Venu M. Nemani, Stuart Hershman, Daniel G. Kang and Michael P. Kelly
The authors examined the correlation between lumbar spine CT Hounsfield unit (HU) measurements and bone mineral density measurements in an adult spinal deformity (ASD) population.
Patients with ASD were identified in the records of a single institution. Lumbar CT scans were reviewed, and the mean HU measurements from L1–4 were recorded. Bone mineral density (BMD) was assessed using femoral neck and lumbar spine dual-energy x-ray absorptiometry (DEXA). The number of patients who met criteria for osteoporosis was determined for each imaging modality.
Forty-eight patients underwent both preoperative DEXA and CT scanning. Forty-three patients were female and 5 were male. Forty-seven patients were Caucasian and one was African American. The mean age of the patients was 62.1 years. Femoral neck DEXA was more likely to identify osteopenia (n = 26) than lumbar spine DEXA (n = 8) or lumbar CT HU measurements (n = 6) (p < 0.001). There was a low-moderate correlation between lumbar spine CT and lumbar spine DEXA (r = 0.463, p < 0.001), and there was poor correlation between lumbar spine CT and femoral neck DEXA (r = 0.303, p = 0.036).
Despite the opportunistic utility of lumbar spine CT HU measurements in identifying osteoporosis in patients undergoing single-level fusion, these measurements were not useful in this cohort of ASD patients. The correlation between femoral neck DEXA and HU measurements was poor. DEXA assessment of BMD in ASD patients is essential to optimize the care of these complicated cases.
Shivanand P. Lad, Raphael Guzman, Michael E. Kelly, Gordon Li, Michael Lim, Karl Lovbald and Gary K. Steinberg
✓Vasospasm following cerebral aneurysm rupture is one of the most devastating sequelae and the most common cause of delayed ischemic neurological deficit (DIND). Because vasospasm also is the most common cause of morbidity and mortality in patients who survive the initial bleeding episode, it is imperative not only to diagnose the condition but also to predict which patients are likely to become symptomatic. The exact pathophysiology of vasospasm is complex and incompletely elucidated. Early recognition of vasospasm is essential because the timely use of several therapeutic interventions can counteract this disease and prevent the occurrence of DIND. However, the prompt implementation of these therapies depends on the ability to predict impending vasospasm or to diagnose it at its early stages.
A number of techniques have been developed during the past several decades to evaluate cerebral perfusion, including positron emission tomography, xenon-enhanced computed tomography, single-photon emission computed tomography, perfusion- and diffusion-weighted magnetic resonance imaging, and perfusion computed tomography. In this article, the authors provide a general overview of the currently available perfusion imaging techniques and their applications in treating vasospasm after a patient has suffered a subarachnoid hemorrhage. The use of cerebral perfusion imaging techniques for the early detection of vasospasm is becoming more common and may provide opportunities for early therapeutic intervention to counteract vasospasm in its earliest stages and prevent the occurrence of DINDs.
Varun R. Kshettry, Michael L. Kelly, Benjamin P. Rosenbaum, Andreea Seicean, Lee Hwang and Robert J. Weil
Myelomeningocele repair is an uncommonly performed surgical procedure. The volume of operations has been decreasing in the past 2 decades, probably as the result of public health initiatives for folate supplementation. Because of the rarity of myelomeningocele, data on patient or hospital factors that may be associated with outcome are scarce. To determine these factors, the authors investigated the trends in myelomeningocele surgical repair in the United States over a 23-year period and examined patient and hospital characteristics that were associated with outcome.
The Nationwide Inpatient Sample database for 1988–2010 was queried for hospital admissions for myelomeningocele repair. This database reports patient, hospital, and admission characteristics and surgical trends. The authors used univariate and multivariate logistic regression to assess associations between patient and hospital characteristics and in-hospital deaths, nonroutine discharge, long hospital stay, and shunt placement.
There were 4034 hospitalizations for surgical repair of myelomeningocele. The annual volume decreased since 1988 but plateaued in the last 4 years of the study. The percentages of myelomeningocele patients with low income (30.8%) and Medicaid insurance (48.2%) were disproportionately lower than those for the overall live-born population (p < 0.0001). More operations per 10,000 live births were performed for Hispanic patients (3.2) than for white (2.0) or black (1.5) patients (p < 0.0001). Overall, 56.6% of patients required shunt placement during the same hospital stay as for surgical repair; 95.0% of patients were routinely discharged; and the in-hospital mortality rate was 1.4%. Nonwhite race was associated with increased in-hospital risk for death (OR 2.8, 95% CI 1.2–6.3) independent of socioeconomic or insurance status.
Overall, the annual surgical volume of myelomeningocele repairs decreased after public health initiatives were introduced but has more recently plateaued. The most disproportionately represented populations are Hispanic, low-income, and Medicaid patients. Among nonwhite patients, increased risk for in-hospital death may represent a disparity in care or a difference in disease severity.
Kelly Wright, Polly Young, Cristina Brickman, Teena Sam, Neeraj Badjatia, Marcus Pereira, E. Sander Connolly and Michael T. Yin
The authors evaluated the rates of ventriculostomy-related infections (VRIs) after antibiotic-coated extraventricular drains (ac-EVDs) were introduced as the standard of care.
A retrospective chart review was conducted of adult patients admitted to NewYork-Presbyterian Hospital neurological intensive care unit in whom an EVD was placed between February 2007 and November 2009, excluding individuals receiving EVDs due to an infection of a primary device. Three time periods were defined depending on type of EVD in use: Period 1, conventional EVDs; Period 2, either ac-EVDs or conventional EVDs; and Period 3, ac-EVDs. Definite/probable VRIs that occurred during the 3 periods were evaluated and established as determinants of VRIs by using a Cox proportional hazards model. Prolonged systemic antibiotics were given for the duration of EVD placement in each of the 3 periods per institutional policy.
Data from 141 individuals were evaluated; mean patient age was 53.8 ± 17.2 years and 54% were female. There were 2 definite and 19 probable VRIs. The incidence of definite/probable VRI (per 1000 person-catheter days) decreased from Period 1 to 3 (24.5, 16.2, and 4.4 in Periods 1, 2, and 3, respectively; p < 0.0001). Patients with VRIs were more likely to be female than male (23.7% vs 3.1%, p < 0.003) and have had an EVD in place for a longer duration, although there was no significant difference among the 3 periods (7.9 ± 6.7 [Period 1], 8.1 ± 7.1 [Period 2], and 8.6 ± 5.8 [Period 3] mean days; p = 0.87, ANOVA). Analysis of effect modification in a stepwise model showed that period, age, and age and female interaction were significant predictors of VRIs. The period was the strongest predictor of VRI (p = 0.0075). After adjustment for age and age and sex interaction, the survival rate was 53% at the end of Period 2 and 91% at the end of Period 3.
Rates of VRIs have decreased with the addition of ac-EVDs to the routine use of prolonged systemic antibiotics at the authors' institution.
Michael Opoku-Darko, Stefan T. Lang, James Artindale, J. Gregory Cairncross, Robert J. Sevick and John J. P. Kelly
Occasionally, diffusely infiltrating low-grade gliomas (LGGs) are identified as incidental findings in patients who have no signs or symptoms that can be ascribed to the tumors. The diagnosis of incidental, asymptomatic LGGs has become more frequent due to the vast increase in access to medical imaging technology. While management of these lesions remains controversial, early surgery has been suggested to improve outcome. The authors set out to identify and review the characteristics and surgical outcomes of patients who underwent surgical intervention for incidental LGG.
All cases of LGG surgically treated between 2004 and 2016 at the authors’ institution were analyzed to identify those that were discovered incidentally. Patients with incidentally discovered LGGs were identified, and their cases were retrospectively reviewed. An “incidental” finding was defined as an abnormality on imaging that was obtained for a reason not attributable to the glioma, such as trauma, headache, screening, or research participation. Kaplan-Meier analysis was performed to determine actuarial rates of overall survival, progression-free survival, and malignant progression–free survival.
In 34 (6.8%) of 501 adult patients who underwent surgery for LGG, the tumors were discovered incidentally. Headache (26%, n = 9) and screening (21%, n = 7) were the most common indications for brain imaging in this group. Four of these 34 patients had initial biopsy after the tumor was identified on imaging. In 5 cases, the patients opted for immediate resection; the remaining cases were managed with a “watch-and-wait” approach, with intervention undertaken only after radiological or clinical evidence of disease progression. The mean duration of follow-up for all 34 cases was 5 years. Twelve patients (35.3%) had disease progression, with an average time to progression of 43.8 months (range 3–105 months). There were 5 cases (14.7%) of malignant progression and 4 deaths (11.8%). Oligodendroglioma was diagnosed in 16 cases (47%) and astrocytoma in 15 (44%). Twenty-five patients (74%) had IDH1 mutation and demonstrated prolonged survival. Only 2 patients had mild surgery-related complications, and 16 patients (47%) developed epilepsy during the course of the disease.
In this retrospective analysis of cases of incidentally discovered LGGs, the tumors were surgically removed with minimal surgical risk. In patients with incidental LGGs there is improved overall survival relative to median survival for patients with symptomatic LGGS, which is likely attributable to the underlying favorable biology of the disease indicated by the presence of IDH1 mutation in 74% of the cases.
Adam P. Burdick, Hubert H. Fernandez, Michael S. Okun, Yueh-Yun Chi, Charles Jacobson and Kelly D. Foote
Adverse event (AE) rates for deep brain stimulation (DBS) are variable, due to various methodologies used for identifying, collecting, and reporting AEs. This lack of a prospective, standardized AE collection method is a shortcoming in the advancement of DBS. In this paper the authors disclose the standardized and prospectively recorded AE data from their institution, correlated with clinical outcome and quality of life (QOL) measures.
All patients who underwent operations at the authors' institution for Parkinson disease (PD), essential tremor, dystonia, other tremor, and obsessive-compulsive disorder were included. Complications occurring intraoperatively or within the first 180 days following surgery were recorded, analyzed, and classified as mild, moderate, or severe, regardless of their perceived relationship to the procedure. The presence, frequency, and severity of AEs were compared with the following outcome measurements: postoperative change in the QOL scales (Medical Outcomes Study 36-Item Short-Form Survey, 39-Item PD Questionnaire); motor scales (Tremor Rating Scale, Unified Dystonia Rating Scale, Unified PD Rating Scale); and Patient Global Impression Scale (PGIS).
Two hundred seventy DBS procedures were performed in 198 patients. Three hundred AEs were recorded in 146 (54.1%) of the 270 procedures, and the AEs were recorded in 119 (60.1%) of 198 patients. Of the 198 patients, the maximum severity of AEs was mild in 28 (14.1%), moderate in 35 (17.7%), and severe in 56 (28.3%). Of the 300 AEs, 102 (34.1%) of 299 were mild, 106 (35.5%) were moderate, and 91 (30.4%) were severe. The AEs were classified as probably not stimulation induced in 10 (3.4%) of 297, probably in 44 (14.9%), unclear for 89 (30%), and not applicable to stimulation in 154 (51.9%). Adverse events were also classified as probably related to surgery in 111 (37.2%) of 298, possibly related in 96 (32.2%), and probably not related to surgery in 91 (30.5%). There was no significant difference (p = 0.22) in QOL outcomes among patients who had no AEs compared with those who experienced mild, moderate, or severe AEs. There was no significant difference in QOL outcomes between patients who did not experience an AE compared with those who experienced any AE. There was no significant difference in the mean General PGIS score between patients without an AE versus those with any AE, as well as on the Symptom-Specific PGIS. Motor function outcomes did not vary between patients with or without AEs. For patients with PD with or without AEs, there was no significant difference in preoperative off-medicine Unified PD Rating Scale score and postoperative 6-month on-medication/on-stimulation change scores (p = 0.59). For patients with tremor there were no differences between those with or without AEs on the Tremor Rating Scale for motor function or activities of daily living. Patients with dystonia with and without AEs showed no differences in the Unified Dystonia Rating Scale.
Prospectively and systematically recording AEs may result in higher AE rates, but this does not correlate with poorer QOL, motor function, or patient-oriented outcome scores.
Michael P. Kelly, Michael A. Kallen, Christopher I. Shaffrey, Justin S. Smith, Douglas C. Burton, Christopher P. Ames, Virginie Lafage, Frank J. Schwab, Han Jo Kim, Eric O. Klineberg, Shay Bess and the International Spine Study Group
After using PROsetta Stone crosswalk tables to calculate Patient-Reported Outcomes Measurement Information System (PROMIS) Physical Function (PF) and Pain Interference (PI) scores, the authors sought to examine 1) correlations with Scoliosis Research Society–22r (SRS-22r) scores, 2) responsiveness to change, and 3) the relationship between baseline scores and 2-year follow-up scores in adult spinal deformity (ASD).
PROsetta Stone crosswalk tables were used to converted SF-36 scores to PROMIS scores for pain and physical function in a cohort of ASD patients with 2-year follow-up. Spearman correlations were used to evaluate the relationship of PROMIS scores with SRS-22r scores. Effect size (ES) and adjusted standardized response mean (aSRM) were used to assess responsiveness to change. Linear regression was used to evaluate the association between baseline scores and 2-year follow-up scores.
In total, 425 (425/625, 68%) patients met inclusion criteria. Strong correlations (all |r| > 0.7, p < 0.001) were found between baseline and 2-year PROMIS values and corresponding SRS-22r domain scores. PROMIS-PI showed a large ES (1.09) and aSRM (0.88), indicating good responsiveness to change. PROMIS-PF showed a moderate ES (0.52) and moderate aSRM (0.69), indicating a moderate responsiveness to change. Patients with greater baseline pain complaints were associated with greater pain improvement at 2 years for both SRS-22r Pain (B = 0.39, p < 0.001) and PROMIS-PI (B = 0.45, p < 0.001). Higher functional scores at baseline were associated with greater average improvements in both SRS-22r Activity (B = 0.62, p < 0.001) and PROMIS-PF (B = 0.40, p < 0.001).
The authors found strong correlations between the SRS-22r Pain and Activity domains with corresponding PROMIS-PI and -PF scores. Pain measurements showed similar and strong ES and aSRM while the function measurements showed similar, moderate ES and aSRM at 2-year follow-up. These data support further exploration of the use of PROMIS–computer adaptive test instruments in ASD.