INTRODUCTION
Cauda equina syndrome (CES) is a clinically defined syndrome characterized by at least one of the following: bladder or bowel dysfunction, reduced sensation in the saddle area, or sexual dysfunction.1 It is a surgical emergency typically caused by a lumbar disc herniation compressing the lumbosacral nerve roots.2 Lumbar stenosis may also cause CES, especially as narrowing of the spinal canal lowers the threshold for nerve compression with smaller disc herniations.3, 4 Low back pain and sciatica (ie, radiculopathy) are common symptoms of CES, each present in >90% of patients.5 CES generally affects adults, with a mean age of 50 years,4 and has an incidence of 270 per 100,000 (0.27%) per year among those with low back pain.6
Patients with early signs of CES occasionally present to direct‐access clinicians (ie, those whom patients can see without a referral) who manage low back pain, such as primary care physicians,5, 7 chiropractors,8, 9, 10 and physical therapists.11 Because chiropractors frequently treat patients with low back pain or radiculopathy12 by using hands‐on manipulation of the spinal joints (ie, spinal manipulation therapy [SMT]), concern has been raised regarding an association between SMT and CES. Chiropractors are trained to recognize the clinical features of CES and refer suspected patients for emergency surgical care.13 However, CES may be preceded by mild initial symptoms such as low back pain for a median of 30 days.14, 15 Additionally, only 27% of patients with CES initially present with more overt symptoms of loss of bowel or bladder function,16 making this condition challenging to recognize even among trained clinicians. This dilemma underscores the need to investigate whether SMT could precipitate CES in patients with predisposing lumbar spine disorders who may have subtle CES‐related symptoms.
Previous case reports and medicolegal cases have described CES after SMT, raising concerns that SMT could trigger CES in predisposed patients.17, 18 Despite this, there is a paucity of high‐level evidence examining the potential SMT–CES association.19 Observational studies and randomized controlled trials focusing on patients with lumbar disc herniation and/or radiculopathy have supported the safe use of SMT for sciatica without causing CES.20, 21, 22, 23 However, trials may be underpowered to examine this potential association and may include patients with less severe or less acute symptoms than those encountered in real‐world clinical practice.20, 21, 22, 23 Chiropractors are often the first clinicians to see patients with disc herniation, radiculopathy, or stenosis,12 necessitating additional research focused on this population.
Our recent retrospective cohort study including 134,440 propensity‐matched U.S. adults with low back pain found no significant increased risk of CES following SMT compared to those undergoing an examination by a physical therapist.19 Although these findings are reassuring, they may not generalize to patients at greater risk of CES. Specifically, in our previous study only ∼1% of patients had lumbar disc herniation with radiculopathy at baseline, 4% had sciatica, and 3% had lumbar stenosis.19 Accordingly, the present study aims to build upon these findings by focusing exclusively on a population predisposed to CES, requiring all patients have either lumbar disc herniation, stenosis, or sciatica/radiculopathy at baseline. This study also builds on the prior methods by (1) requiring physical therapist treatment (ie, exercise) rather than evaluation alone to reduce bias from potential baseline CES suspicion; (2) examining additional CES‐related outcomes of bladder catheterization and fecal incontinence; (3) applying a broader set of exclusions (eg, rare CES etiologies, functional neurological disorders, recent spine surgery) to improve cohort homogeneity; and (4) incorporating over a year of updated data.
Given the limited research on the topic to date, we aimed to investigate the association between SMT and CES in adults with low back disorders including disc herniation, stenosis, and/or radiculopathy. In this population, we tested the null hypothesis that SMT is not associated with a statistically significant positive risk ratio (RR) of CES compared to matched controls receiving physical therapist‐led therapeutic exercise (PTE), over 3 months follow‐up. Secondary outcomes explored the cumulative incidence of CES, and, because bowel and bladder dysfunction are commonly associated with CES, we also calculated the RRs for bladder catheterization and bowel incontinence.
MATERIALS AND METHODS
Study design
The present study used data from TriNetX (TriNetX, LLC, Cambridge, MA, USA), a federated research network platform that provides access to aggregated electronic health record data from a network of 103 large health care organizations covering >141 million patients at the time of our query. With a database query date of January 13, 2025, we included patients who met eligibility criteria anywhere from January 13, 2025, to October 13, 2025 (spanning 20 years to 3 months prior to the query date to allow for adequate follow‐up). This data range was selected to maximize sample size while considering database limitations.
Specifically, TriNetX limits analyses to patients meeting eligibility criteria a maximum of 20 years retrospectively.
The platform allows users to conduct queries to analyze patient cohorts, using standardized nomenclatures such as the International Classification of Diseases, Tenth Revision (ICD‐10).24, 25 These codes are automatically interconverted to Ninth Revision (ICD‐9) as needed. TriNetX conducts data quality procedures to ensure cleanliness, consistency, correctness, and completeness of data.25 All data available via the platform are deidentified in accordance with Health Insurance Portability and Accountability Act guidelines. A visual representation of the study design is shown in Supplemental File 1 Figure S1.
Study methods adhere to a registered protocol,26 and reporting follows Strengthening the Reporting of Observational Studies in Epidemiology.27
The first author's institutional review board (IRB) considers studies using deidentified data from the online TriNetX platform Not Human Subjects Research thereby exempting the present study from institutional review board review and waiving the need for consent.
A key feature of this study is the use of PTE as an active comparator.28 Further distinguishing this study from the previous one focused on CES risk after SMT,19 the present study explicitly requires patients in the comparator PTE cohort to undergo therapeutic exercise, such that both cohorts receive active care. This design feature aims to make the cohorts more comparable with respect to their complexity and care eligibility. Like chiropractors, physical therapists are also direct‐access clinicians who commonly manage low back pain.29 Between 36% and 40% of patients with lumbar disc herniation visit a physical therapist30, 31 and 16% receive SMT administered by a chiropractor,31 suggesting that both cohorts reflect common care pathways for low back disorders.
Patients who visit physical therapists for spinal pain are generally more comparable to chiropractic patients than those visiting medical physicians or specialists with respect to socioeconomic factors, insurance coverage, comorbidity burden, and self‐rated health status.32 Several commonly used therapeutic exercises such as general and motor control exercises have evidence supporting their use for lumbar disc herniation.33, 34 Finally, PTE is not associated with an increased risk of CES to our knowledge.33
Participants
Eligibility criteria
We included adults at least 18 years old presenting with a low back disorder of lumbar disc herniation, sciatica/radiculopathy, and/or lumbar stenosis (Supplemental File 1 Table S1).
Patients were divided into two cohorts based on either receiving chiropractic SMT (Current Procedural Terminology [CPT]: 98940, 98941, or 98942) or PTE (CPT codes: 97001, 97161, 97162, or 97163 [physical therapy evaluations] and 97110 [therapy procedure using exercise]). Patients were identified at the first co‐occurrence of an eligible lumbar spine disorder (ie, lumbar disc herniation, sciatica/radiculopathy, or stenosis) and SMT or PTE, which served as the index date. This helped avoid prevalent users of SMT or PTE for these disorders for whom CES screening and/or decision‐making could be less impactful on care delivery.
To help ensure data completeness, we required patients to have a previous health care visit between 1 day and 3 years preceding the index date (ie, the date when patients received SMT or PTE and were included in the study). To minimize loss to follow‐up, we required at least one health care visit between 1 day and 3 months of follow‐up.
In addition to the requirement of having a CPT code for SMT administered by a chiropractor, patients receiving SMT were required to have the presence of a segmental dysfunction code for the thoracic or lumbopelvic regions (ie, M99.02, M99.03, M99.04, M99.05) indicating that SMT was applied to any of these regions.18 Consequently, patients receiving only cervical SMT were not included.
We excluded patients with preexisting CES, cauda equina injury, spinal cord injury, congenital abnormalities of the cauda equina, conus medullaris syndrome, urinary or fecal incontinence, bladder catheterization,15 serious lumbar spinal pathology (ie, fracture, malignancy, infection, and bleeding disorders) that may cause CES,1, 3 recent spine surgery or injection,35, 36 those with functional neurological disorders or simulated illness that may mimic CES, and individuals with rare etiologies of CES (eg, nervous system Zoster, Guillain–Barré syndrome)37, 38 (Table S2). These exclusions aimed to make the cohorts more homogeneous with respect to their low back conditions. Those receiving manual therapy (CPT: 97140), chiropractic SMT (CPT: 98940, 98941, 98942), or osteopathic manipulation (CPT: 1013558) were excluded from the PTE cohort.
This strategy aimed to minimize exposure misclassification considering some physical therapists may provide SMT,39 or patients could receive SMT concurrently from osteopaths or chiropractors surrounding the index date. Exclusions are detailed in Supplemental File 1 Table S2.
Variables
We used propensity score matching to minimize bias by balancing variables associated with CES,28 including age, gender, comorbidities, degenerative lumbar spine conditions that may cause CES (eg, disc herniation, spondylolisthesis, stenosis), other CES risk factors such as trauma and neurological disorders, and medications used for low back pain that increase the likelihood of urinary retention.37, 38, 40, 41, 42 Variables available within 1 year preceding the index date were eligible for matching (Supplemental File 1 Table S3).
Primary outcome
We ascertained new instances of CES (ICD‐10: G83.4) over a 3‐month window including and following the index date. Previous studies established that although the median time to diagnose CES is <2 weeks, patients may have symptoms up to 3 months or more before being diagnosed.43, 44, 45, 46 Shorter time windows (eg, 0 and 1 day) were avoided as these may miss a substantial number of CES diagnoses. Longer windows were avoided considering this would allow for additional confounding due to other intervening factors after the index date (eg, discontinuation of treatment or use of other therapies).
We presented findings in the unmatched sample as a form of sensitivity analysis.
Secondary outcomes
We analyzed postmatching secondary outcomes apart from CES considering that clinicians may append the diagnosis using variable thresholds.2, 5, 47 Accordingly, secondary outcomes were used to determine the robustness of our primary outcome. We examine the RR for new instances of sequelae of CES including urinary catheterization and fecal incontinence.5
We further characterized the SMT cohort according to the mean number of follow‐up SMT visits (CPT: 98940, 98941, or 98942). We reported and compared the proportion of patients receiving physical therapy‐related intervention(s) or evaluation(s) between cohorts and reported the number of follow‐up physical therapy visits among patients with nonzero counts of these visits using the mean, SD, and median visit count. A range of codes were used to capture these encounters (Supplemental File 1 Table S4).
Statistical methods
We conducted statistical analysis using features within the TriNetX database analytics platform. To compare the baseline characteristics, we used standardized mean deviation (SMD) with a threshold of >0.1 indicating between‐cohort imbalance. Propensity score matching was performed using a greedy nearest‐neighbor approach with a 1:1 ratio and a caliper of 0.1 pooled standard deviations.
A logistic regression was applied using Python (scikit‐learn version 1.3 [Python Software Foundation, Delaware, USA]) to pooled covariate matrices to predict the probability of each patient being in the control cohort. The matching process cycles through each patient in the smaller cohort and identifies the closest unmatched patient from the larger cohort, excluding non‐matching patients.
We used R (build 4.3.2, Vienna, Austria48) to calculate RRs with 95% confidence intervals (CIs), calculate risk difference (RD) for CES and its 95% CI using the Miettinen and Nurminen method, using DescTools,49 and derived p values for RRs and risk difference using the chi‐square test with a significance threshold of p < .05. We also used ggplot250 to plot CES incidence and cumulative incidence per cohort.
We took additional steps to explore data quality and balance diagnostics. This included reporting measures of cohorts' follow‐up duration, the proportion of unknown variables per cohort, and plots of covariate balance and propensity score density using ggplot2.50 We explored RRs for negative control outcomes unrelated to SMT51, 52 (Supplemental File 1 Table S5). These were evaluated by assessing whether the majority of these RRs fell within a predefined range surrounding the null (0.73 ≥ RR ≤1.38), suggestive of between‐cohort balance.52, 53 Imbalance would suggest that additional matching may be necessary or that caution is needed when interpreting results.
Study size
We estimated a total required sample of 51,856 using G*Power (Kiel University, DE) z‐tests to examine a difference in incidence proportion between cohorts of 0.015% vs. 0.030%,6 using a two‐tailed α‐error of .05, power of 0.95, and allocation ratio of one. Initial test queries suggested this sample size was attainable.
RESULTS
Participants
Before matching, there were 34,434 patients in the SMT cohort and 630,160 in the PTE cohort. Compared to the PTE cohort, patients in the SMT cohort were initially younger, with a lower proportion who had been prescribed opioids, benzodiazepines, or gabapentinoids and lower incidence of external causes of morbidity (eg, motor vehicle collisions), among other differences (SMDs >0.1; Table 1). After matching, there were 34,376 patients in each cohort and all variables were adequately matched, having SMD values <0.1.
A patient selection flowchart is shown in Figure S2.
Data quality
After matching, the propensity score densities of each cohort were superimposed (Supplemental File 1 Figure S3), and all SMD values were below the threshold for balance (Table 1; Supplemental File 1 Figure S4). The proportion of patients with an unknown gender was approximately 0% in both cohorts (SMD = 0.048). Comparing the SMT to PTE cohort over the 90‐day follow‐up, the RRs for negative control outcomes were in the target zone for balance.
This included colonoscopy (RR = 1.28 [95% CI, 1.14–1.44]), azithromycin prescription (RR = 1.13 [95% CI, 1.04–1.23]), and acute upper respiratory infection (RR = 1.21 [95% CI, 1.13–1.30]). The proportion of patients who had data spanning at least the 90‐day follow‐up was high in each cohort (SMT: 96.7%; PTE: 94.5%), and the mean and median follow‐up duration were also similar between cohorts (Supplemental File 1, Figure S5 and Table S6). Accordingly, these findings suggest the presence of adequate and similar markers of data completeness, indicate that propensity matching was successful, and suggest the absence of differential attrition between cohorts.
Primary outcome
Following matching, the total number of CES cases was 14 in the SMT cohort and 16 in the PTE cohort. Comparing the SMT cohort to PTE cohort, the incidence and risk of CES did not significantly differ (0.04% vs. 0.05%; RR = 0.88 [95% CI, 0.43–1.79]; p = .715), as presented in Table 2 and Figures 1 and 2.
The RD further suggested no meaningful difference in risk of CES (RD = −0.01% [95% CI, −0.04% to 0.03%]; p = .715).
Secondary outcomes
Following matching, comparing the SMT cohort to the PTE cohort, the incidence and risk of bladder catheterization (RR = 0.50 [95% CI, 0.39–0.64]; p < .001) and fecal incontinence (RR = 0.50 [95% CI, 0.37–0.68]; p < .001) were significantly lower, as presented in Table 3. Total and cumulative incidences for these outcomes are demonstrated in Supplemental File 1 Figures S6– S9.
Patients receiving SMT received a mean (SD) 6.2 (5.5) sessions of SMT during the 90‐day follow‐up window, yielding an estimated total of 213,131 SMT sessions for the entire cohort. In the SMT cohort, 15,008 patients received physical therapy‐related intervention(s) or evaluation(s), and 22,552 patients in the PTE cohort received these therapies during the 90‐day follow‐up. Accordingly, the incidence and likelihood of receiving any physical therapy intervention or evaluation were significantly lower in the SMT cohort compared to PTE cohort (43.66% vs.
65.60%; RR = 0.67 [95% CI, 0.66–0.68]; p < .0001). However, among individuals receiving physical therapy‐related follow‐up visits, the mean count of such visits was slightly greater in the SMT cohort [SD] (mean = 5.7 [5.2]) versus PTE cohort (mean = 5.1 [4.7]) with both cohorts having a median of four visits.
DISCUSSION
This study found that CES risk does not increase after SMT compared to PTE among adults with lumbar disc herniation, radiculopathy, or stenosis. Our use of propensity score matching helps ensure that these findings were independent of recognized confounders. The cumulative incidence data further support that CES occurs at similar rates in the study population, regardless of treatment received being SMT or PTE.
Our secondary outcomes corroborate these results, showing no significant increase in CES‐related events such as bladder catheterization or bowel incontinence among SMT recipients. Additionally, our data suggest that PTE is relatively safe considering the low overall incidence of CES, which was not significantly different between cohorts after matching.
Direct comparisons of our observed CES incidence with previous incidence estimates are precluded by differences in study populations and methods. However, there are similarities in the present CES incidence with the general low back pain population, rather than a potential increase related to our focus on disc herniation, radiculopathy, and stenosis. The incidence estimates for CES in our study, which range from 0.04% to 0.06% across both cohorts, align with a prior study indicating that CES affects 270 per 100,000 (0.27% [95% CI, 0.14%–0.54%]) annually among adults with low back pain presenting to secondary care settings,6 translating to an incidence of approximately 0.07% per 90 days.
One explanation of the similar CES incidence values between our study, which focused on patients with lumbar spine disorders, and prior studies focusing on a more general low back pain population may relate to our exclusion of patients with other neurological disorders. These exclusions likely reduced our observed CES incidence by avoiding ascertainment of scan‐negative CES.5, 37, 38 Another explanation for the similar incidence is that patients with more severe initial presentations of lumbar spine disorders could have avoided seeking SMT or PTE and instead presented to an emergency setting6 and were not included in our study. Similarly, any patients for whom chiropractors or physical therapists suspected CES and referred for emergency surgical evaluation rather than providing treatment have been excluded, thereby counterbalancing any potential increase in observed CES incidence in this population.
Our eligibility criteria aimed to reflect real‐world chiropractic practice by excluding patients with serious spinal pathologies who would not typically see a chiropractor and for whom SMT is contraindicated (eg, those with recent spine fracture, spinal infection, and known CES‐related conditions).54, 55, 56, 57 This selective narrowing likely excluded patients with high CES risk. However, our design focused on patients with lumbar disc herniation, sciatica/radiculopathy, and/or stenosis, which represent the primary etiologies or risk factors for CES.2, 3, 4, 5 Chiropractors regularly use SMT for these lumbar spine disorders in the absence of overt CES features.12 Accordingly, our study design allowed us to test the potential risk association of SMT in a clinically relevant population.
The findings suggesting a significantly reduced likelihood of bladder catheterization and fecal incontinence among SMT recipients should be interpreted with caution. These outcomes are not specific to CES and may arise from various benign reasons. Chiefly, patients receiving PTE may have had undiagnosed or concurrent genitourinary or bowel dysfunction at baseline.
For instance, individuals with interstitial cystitis may experience pelvic floor dysfunction and seek physical therapy, while also undergoing bladder instillation, a procedure that requires catheterization.58 Propensity score matching cannot feasibly control for all variables influencing bladder or bowel function, as doing so would lead to excessive trimming and potentially widen the effect estimate precision for our main outcome of CES.59 Our findings do not imply that PTE contributes to bladder catheterization or fecal incontinence, and instead should be viewed as reinforcing our primary null hypothesis.
The findings of our study are supported by various experimental and observational studies. For instance, randomized clinical trials21, 22 and multiple prospective observational studies20, 60, 61 investigating SMT for lumbar disc herniation or radiculopathy reported no serious adverse events such as CES. One chart review that examined potential adverse events after 960,140 chiropractic SMT sessions identified no cases of CES.62 Additionally, one retrospective study identified a reduction in the likelihood of discectomy among SMT recipients compared to matched controls, serving as a proxy for the safety of SMT considering CES typically necessitates surgical intervention.63 Although biomechanical data on this topic remain limited, one observational study found no significant short‐term changes in disc herniation size or morphology following SMT,64 suggesting that SMT likely does not acutely worsen disc herniations.
Our study helps contextualize the medicolegal and case reports documenting CES following spinal SMT,17, 18 suggesting caution in their interpretation due to inherent publication bias and the absence of epidemiological methods or control groups. SMT and PTE are common entry points for lumbar spine disorders,29, 30, 31, 65 and CES may occur more often in this population regardless of treatments rendered. Accordingly, SMT or PTE may have been incidentally performed while CES was developing, as may initially present with early or benign‐appearing symptoms.16
Another explanation for the observed absence of increased risk of CES after SMT is that chiropractors may tailor SMT techniques to the specific lumbar spine disorder(s) present (ie, lumbar disc herniation, radiculopathy, and/or stenosis). A range of SMT techniques is available, spanning from low to high force, and including both thrust and nonthrust varieties. One example is flexion‐distraction, a technique involving nonthrust mobilization and manual traction.
A survey of 280 chiropractors indicated that although only 29% of respondents reported using flexion‐distraction for lumbar disc herniation with radiculopathy, this technique was used more frequently for this condition compared to other diagnoses, such as facet syndrome and myofascial pain.66 However, thrust manipulation was still commonly used in these patients, with approximately 53% of respondents indicating they would use it in such cases.66 Regardless, the established variety of approaches, combined with the lack of observed risk for CES, may indicate proficiency in managing lumbar spine disorders. Nevertheless, it remains essential for chiropractors to conduct thorough examinations and remain vigilant in assessing patients for CES, ensuring timely referral for emergency surgical evaluation when necessary.13, 56
Existing care guidelines recommend SMT either as a stand‐alone intervention67 or as part of a broader multimodal strategy68, 69, 70, 71, 72 for patients with lumbar disc herniation and/or radiculopathy. The evidence supports the relative safety of SMT within this population, suggesting that SMT can be a viable treatment option in the absence of clinical features of CES or progressive motor loss.73, 74 Although clinical guidelines have found insufficient evidence for use of SMT for lumbar stenosis,74 this condition appears to have made up only a small proportion of the current study population. Recent evidence regarding the utility of SMT in lumbar stenosis has been promising,75 although additional studies devoted to the safety and effectiveness of SMT are needed.
Questions remain regarding the mechanisms underlying the observed null SMT–CES association. For instance, the present study does not elucidate whether our findings reflect the inherent safety of SMT itself, the clinical decision‐making processes of chiropractors, or a combination of both. Future research including more granular data extracted from clinical charts could examine how chiropractors and physical therapists manage patients with lumbar disc herniation, radiculopathy, and stenosis.
Such studies could investigate examination thoroughness and diagnosis and referral patterns of CES, potential associations between specific SMT techniques and CES, or patient selection criteria when determining whether to administer SMT or exercise. Additional surveys exploring how chiropractors might modify treatments for patients with lumbar disc herniation would also be valuable.66 Finally, given that CES is rare but serious, developing prediction models to identify which patients presenting for SMT or PTE may be at highest risk could facilitate earlier triage and surgical evaluation. Such studies could incorporate regression models with predictors including demographics, comorbidities, body mass index, and clinical variables such as pain severity, disability, or specific examination findings.
STRENGTHS AND LIMITATIONS
The present study strengths include an interdisciplinary investigator team, a priori protocol,26 and methodological improvements compared to the previous study on this topic including19 (1) focus on a population with a greater baseline risk of CES, (2) use of a more active comparator undergoing PTE, (3) a broader set of exclusions, (4) more comprehensive propensity matching strategy, (5) use of outcome negative controls, and (6) addition of secondary outcomes.
Several limitations are noteworthy. The relatively low number of observed CES events per cohort, yielding wide 95% CIs for incidence estimates, reflects the rarity of CES and limits precision of our effect estimates. This warrants some caution in interpretation, even though the study was adequately powered via a priori calculation, with >68,000 patients, and corroborates a previous study.19 A follow‐up study with a larger sample size could be conducted using the TriNetX Linked Network, which links electronic health records with claims data, including >1.5 million chiropractic patients (vs. ~150,000 in the current Research Network).24, 76, 77 A key factor leading to our observed imprecision is the comparatively small chiropractic cohort size, whereas the larger PTE cohort produced narrower incidence estimates prior to matching.
A follow‐up study using TriNetX Linked would require some adaptation of the present methods, for instance a narrowing of the data range to match the range in Linked (approximately 15 years). Overall, this could yield up to a 10‐times larger chiropractic cohort size, improved precision, and potentially greater generalizability to broader private practice chiropractic settings.
There is a lack of definitive diagnostic criteria for CES, and there is a potential for scan‐negative CES.5, 37, 38 Although the potential impact of false‐positive CES diagnosis is mitigated by our selection criteria and matching strategy, an artificial increase in CES incidence in one or both cohorts remains possible. Regardless, we have no reason to suspect that documentation bias would differ between cohorts, which were both engaged in large health care organizations. We are unable to compare our query against chart review given the deidentified nature of the dataset.
There may be unmeasured confounding related to between‐cohort differences related to patients' baseline pain severity, disability level, or symptom distribution due to limitations in ICD‐10 granularity. Furthermore, we were unable to determine biomechanical or technique parameters of SMT used, such as the type of thrust, patient positioning, or force‐time characteristics, due to limited data granularity and an inability to observe patient‐level information. In addition, we could not determine whether SMT procedures were modified in response to comorbidities or precautions such as lumbar stenosis or osteoporosis.62, 66 We were also unable to determine the specific types of therapeutic exercises administered in the PTE cohort.
Finally, the findings may not generalize to other countries wherein the incidence and treatment pathways for CES may vary.
CONCLUSION
The present study identified no increase in risk of CES or secondary markers of this condition among adults with lumbar spine disorders receiving SMT compared to matched controls receiving PTE. These findings suggest that the likelihood of CES following chiropractor‐administered SMT is similar to that following exercise administered by a physical therapist in this population. Based on an adequately powered sample of 68,752 patients, these results are reassuring and corroborate previous research on the topic.
Replication using a larger dataset could enhance the precision of CES incidence estimates and improve the certainty of findings.
FUNDING INFORMATION
This project is supported by the Clinical and Translational Science Collaborative of Northern Ohio which is funded by the National Institutes of Health, National Center for Advancing Translational Sciences, Clinical and Translational Science Award grant, UM1TR004528. The work conducted by R.T. received support from the Elisabeth Severance Prentiss Foundation (Cleveland, OH) through general funding. The study did not receive a specific grant, and the funders were not involved in developing the study protocol or any key decisions.
ETHICS STATEMENT
The University Hospitals Institutional Review Board (Cleveland, OH, USA) considers studies using deidentified data from the online TriNetX platform to be Not Human Subjects Research thereby exempting the present study from institutional review board review and waiving the need for consent.
DISCLOSURE
Robert J. Trager acknowledges that he has received royalties as the author of two texts on the topic of sciatica. The other authors have declared no competing interests.
This journal‐based CME activity is designated for 1.0 AMA PRA Category 1 CreditTM. Effective January 2024, learners are no longer required to correctly answer a multiple‐choice question to receive CME credit. Completion of an evaluation is required, which can be accessed using this link, https://onlinelearning.aapmr.org/.
This activity is FREE to AAPM&R members and available to nonmembers for a nominal fee. CME is available for 3 years after publication date. For assistance with claiming CME for this activity, please contact (847) 737–6000.
All financial disclosures and CME information related to this article can be found on the Online Learning Portal (https://onlinelearning.aapmr.org/) prior to accessing the activity.
Supporting information