APPLYING EVIDENCE TO INDIVIDUAL CASE MANAGEMENT: AN ILLUSTRATIVE EXAMPLE OF CONSERVATIVE VERSUS SURGICAL TREATMENT OF THORACIC OUTLET SYNDROME

INTRODUCTION

Thoracic outlet syndrome (TOS) is a nonspecific diagnosis of the upper limb that involves the compression of vascular or neurogenic structures that pass through the thoracic outlet. This region is bordered by the clavicle, scapula, first rib, anterior scalene and middle scalene and comprises three compartments: (i) the interscalene triangle, (ii) costoclavicular space, and (iii) retropectoralis minor space.¹ Each compartment is a potential site of neurovascular compression.¹ The anatomical structures that are most frequently compressed in these areas are the brachial plexus, axillary/subclavian vein, and subclavian artery.¹ Most cases of TOS are neurogenic in origin² and reported symptoms vary. Symptoms of TOS include numbness, tingling, loss of function, and pain throughout the distribution of the affected nerve or vasculature.³ Although TOS can occur in any individual, risk factors include athletics,⁴ repetitive upper limb motions, poor posture, or pregnancy.⁵

There are various treatment options for patients with TOS that include physical/manual therapy, pain management, or surgery.^6,7 In order for the clinician to determine the most appropriate management strategy for an individual case, an evidence-based approach⁸ can be utilized whereby the findings of research studies are critically appraised for their validity, importance, and applicability to the patient (Table 1). This application of research, combined with the clinician’s experience and the patient’s values and preferences, informs a treatment plan that is based on the best available evidence and that is tailored to the individual patient. For a more complete review of the principles of evidence-based practice, particularly in relation to the chiropractic profession, we refer readers to the textbook by Haneline.⁹

At D’Youville University in Buffalo, New York,¹⁰ chiropractic students undertake a series of evidence-based practice courses as a core component of the academic program. These courses enable students to acquire skills in scholarly writing, literature searching, formulation of clinical questions, critical appraisal of the literature, and application of research findings to clinical case management. Students are also required to write and present evidence-based case reports (EBCRs) to further enhance these skills.¹¹ In some instances, the EBCR is based on the management of an actual patient in the clinical setting, while in others, as in the current paper, the EBCR is based on a patient scenario.¹⁰

To date, despite calls for publishing EBCRs in chiropractic journals,¹¹ there are only a few EBCRs that have been published in the peer-reviewed chiropractic literature.^12–14 EBCRs encompass all the steps of an evidence-based approach to patient management (see Table 1).^11,15,16 By incorporating a focused literature search, EBCRs can also highlight potential knowledge gaps in the evidence base for particular clinical topics.¹⁵ As such, several EBCRs have been published in the peer-reviewed medical and allied health literature,^11,16 including in the British Medical Journal since 1998.¹⁵ Unlike traditional case reports,¹⁷ which present new or unusual findings related to the management of a patient, EBCRs are intended to illustrate the practitioner’s process to the patient’s management and show how evidence was applied in answering different types of clinical questions (e.g., etiology, diagnosis, prognosis, therapy, risk, or cost-effectiveness) throughout the various stages of the patient’s care.^11,15,16 This is congruent with an integrated evidence-based approach to health care which promotes the collection, interpretation, and application of research evidence to patient management.¹⁸ An evidence-based approach has been adopted by several health care professions as the standard of practice for improving patient outcomes and facilitating greater interprofessional cooperation.¹⁹

By using a case scenario of TOS as an example, formatted within an EBCR framework,^11,15,16 the purpose of our article is to illustrate how research findings can be utilized in clinical practice to inform the management of an individual patient.

METHODS

Case Scenario

A case scenario of a 29-year-old pregnant woman with a clinical diagnosis of TOS is summarized and presented in Table 2. This scenario evoked a clinical question and literature search to determine whether chiropractic care or surgical intervention would be the best treatment option.

Clinical Question

The clinical question in our case was about therapy, that is, will chiropractic care help our patient, or should she be referred back to her family physician with a recommendation for consideration of surgical consultation? To answer this, the following foreground question^8,9 was posed: In a 29-year-old pregnant woman with a clinical diagnosis of TOS, would conservative therapy be more effective than surgical intervention at reducing her neurogenic symptoms?

Using the mnemonic, PICO, which stands for Patient/problem, Intervention, Comparison, and Outcome, this question can be broken down as follows:

P – 29-year-old pregnant woman with right-sided TOS

I – Conservative therapy

C – Surgical intervention

O – Relief of neurogenic symptoms

Literature Search

The best study design to evaluate the effectiveness or safety of therapeutic interventions is a systematic review of randomized controlled trials (RCTs).^9,20 If a systematic review is not available, the next best level of evidence is a high-quality RCT (Table 3).^9,20 To begin, a search of the English language literature was conducted using the PubMed database. Inclusion criteria consisted of articles published in English, conducted on human subjects, and published within the last five years. A 5-year cut-off was used to capture the most recent literature on our topic. The medical subject headings 'thoracic outlet syndrome’ and 'conservative therapy’ were combined using the Boolean operator 'AND’, yielding 9 results. After filtering the search to 'systematic reviews’ and 'randomized controlled trials’, a single citation was produced. Therefore, one recent RCT, the first of its kind,³ pertaining to the comparison of surgical versus conservative treatment for TOS was identified and retrieved. The entire literature search, including retrieval of the manuscript, took less than 10 minutes.

RESULTS

Critical Evaluation of the Evidence

In the RCT retrieved,³ surgical intervention was found to be most useful for TOS in patients that were non-responsive to preliminary conservative therapy. There were two interventions in this study: (i) transaxillary thoracic outlet decompression (TA-TOD) versus (ii) continued conservative treatment with physical therapy. Follow-up assessments were performed by a neurologist to evaluate subjective findings via the administration of disability questionnaires. Before the results of this study could be applied to the current patient, the article was appraised using a template provided by the Critical Appraisal Skills Program (CASP) for RCTs.²¹ Specifically, this paper was appraised with respect to its (a) validity, (b) importance, and (c) applicability to the treatment of the current patient. The CASP tool was chosen over other critical appraisal instruments, such as the SIGN (Scottish Intercollegiate Guidelines Network) criteria, because, in addition to items for appraising an article’s internal validity, the CASP tool includes items pertaining to the critical appraisal of the importance and applicability of the article’s findings to patient management.

(a) Are the results of this RCT valid?

1. Did the study address a clearly focused research question?

   • The study by Goeteyn et al.³ addressed a clearly focused research question (i.e., effectiveness of surgical versus conservative intervention in reducing neurogenic symptoms in adult patients with TOS). The inclusion and exclusion criteria in the methods of this study ensured that all participants had undergone six months of conservative treatment, were fit for surgery, and were 18 years of age or older. The study intervention was surgery (TA-TOD) and the comparison intervention was continued conservative therapy. Participant outcomes in the study were clearly identified using functional outcome scores.

2. Was the assignment of participants to interventions randomized?

   • The participants were randomly assigned to the treatment and control groups using a computerized randomization procedure. The randomization logs were kept by an independent third party and results were communicated to the coordinating investigator in a secure way to avoid allocation bias.

3. Were the participants, investigators, and outcome assessors/analysts ‘blinded’ to the treatment administered?

   • Due to the nature of the treatment interventions, this was a non-blinded study. Outcome assessors and analysts were also unblinded.³

4. Were the study groups similar at the start of the RCT?

   • The baseline characteristics (Appendix 1) and disability scores (Appendix 2) for each group were clearly identified and comparable, with the exception of baseline symptom duration (see Appendix 1), which was not controlled for by the investigators in their statistical analyses.

5. Were all participants who entered the study accounted for at its conclusion?

   • Losses and exclusions after randomization were less than 20% and accounted for, including one patient that became pregnant and three patients that refused to fill out follow-up questionnaires. All participants were analyzed in the study groups to which they were randomly assigned (i.e., intention-to-treat).

6. Apart from the experimental intervention, did each study group receive the same level of care (i.e., were they treated equally)?

   • Follow-up intervals were the same for each study group, with functional outcome scores measured at 3, 6, and 12 months.

Therefore, based on the above criteria, we deemed the results of this RCT valid.

(b) Are the valid results of this RCT important?

We deemed the results important for several reasons. First, patient outcome scores were measured and calculated using 5 functional outcome questionnaires: (i) the Disability of Arm, Shoulder, and Hand (DASH) questionnaire, which is a validated measure for upper extremity musculoskeletal complaints²²; (ii) the Cervical-Brachial Symptoms Questionnaire (CBSQ); (iii) the TOS disability scale; (iv) the Short Form-12 (SF-12) physical composite scale (PCS); and (v) the SF-12 mental component summary (MCS).³ The statistical tests utilized were the student t-test (or Mann-Whitney U test when appropriate) and repeated measures analysis of variance (ANOVA) with adjustment for multiple comparisons (Bonferroni). Confidence intervals (CIs) and p-values were reported for all follow-up intervals and treatment groups (see Appendix 2). The between-group clinical outcomes were found to be statistically significant (p < 0.05) at three months, favoring surgical treatment, for four out of the five outcome measures (i.e., DASH, CBSQ, TOS disability scale, and SF-12 PCS). On the DASH questionnaire, for example (see Appendix 2), patients in the surgical group experienced a nearly 20% greater improvement in upper extremity function by 3 months compared to those in the physical therapy group (i.e., mean between-group difference = 19.77; 95% CI, 19.46 to 20.09), which also exceeded the minimum clinically important difference.²³ After the third month of the trial, all participants in the conservative treatment group were offered and accepted the surgical intervention. Subsequent between-group analyses of the functional outcome scores at months six and 12 were no longer statistically significant. The repeated measures ANOVA for the complete 12-month follow-up period showed a statistically significant difference in DASH, CBSQ, TOS disability scale, and physical quality of life (SF-12 PCS) scores favoring surgical treatment.

Clinical outcome success and failure in the study were measured at one year (Appendix 3). Failure of surgery in the conservative treatment group showed a greater than 30% change and was therefore clinically significant (conservative treatment: failure at one year = 31.8%). Failure of surgery was not clinically significant in the surgical group (TA-TOD: failure at one year = 8.3%), suggesting that earlier rather than delayed surgical intervention for TOS may be more effective.

(c) Are the valid, important results of this RCT applicable to our patient?

The current patient matches the characteristics of participants in the study by Goeteyn et al.³ in terms of age, sex, and clinical condition (i.e., TOS diagnosis); however, the study investigators included participants that had undergone previous conservative treatment and were experiencing symptoms for at least one year before being enrolled in the trial. Another factor to consider is that the patient in our scenario is pregnant. Exclusion of a participant occurred in the Goeteyn et al.³ trial after one became pregnant. Our patient’s symptoms began soon after becoming pregnant, indicating that her TOS symptomatology may resolve itself after gestation. The study’s findings may therefore be applicable to our patient if she continued to experience symptoms of TOS after gestation and several months of conservative therapy without reduction of symptom severity.

DISCUSSION

Based on our critical appraisal, it was determined that the study by Goeteyn et al.³ provided valid and important information regarding the outcome of TOS decompression surgery and how it compares to continued conservative therapy in improving functional outcome scores in patients with TOS. Conservative therapy for a minimum of three months has generally been regarded in the literature as the first course of treatment for patients with this condition.^1,6,7 Nonoperative therapies for TOS including physical therapy/chiropractic, orthopedic braces, and taping have been identified as effective forms of treatment in patients who are middle-aged, female, obese or have poor posture.¹ These therapies have been shown to reduce neurogenic TOS symptoms (i.e., upper extremity pain and paresthesia), improve function, and facilitate return-to-work.⁶ If there is no response to conservative therapy after 3 months and/or progressive neurological deterioration, surgical intervention is often recommended.^6,7 Findings from retrospective observational studies have suggested that transaxillary resection of the first rib is a safe and effective surgical strategy for patients with TOS.⁷ However, prior to the study by Goeteyn et al.,³ RCTs on the safety and effectiveness of surgical versus conservative treatment interventions for TOS have been lacking. Full descriptions and additional references of the surgical and conservative treatment interventions used in the Goeteyn et al. trial,³ including specific mobilizations, stretching and strengthening exercises for patients undergoing conservative treatment, are provided by the study authors.³

Application of the Evidence

By using the evidence provided in the RCT by Goeteyn et al.,³ combined with our patient’s clinical state and circumstances, we would advise our patient to receive conservative treatment throughout the duration of her pregnancy and for six months after. If at that point the patient was still experiencing TOS symptoms, a decreased quality of life and/or progressive neurological deficits (e.g., increasing hand-grip weakness or muscle atrophy), the patient would be referred back to her family physician, where referral to an orthopedic surgeon for surgical consultation may be warranted. However, in line with an evidence-based approach,^8,9 the patient’s preferences would also be taken into consideration before any invasive treatment plans were made.

Explanation of the study’s findings to the patient would involve describing the statistically significant difference in the functional outcome scores favoring participants that underwent surgical intervention earlier in the trial.³ Extrapolating the findings to our patient, we can be 95% confident that she would also be expected to benefit by a clinically meaningful amount with surgery, at least in terms of her upper extremity function. Failure rates at the 1-year follow-up in the Goeteyn et al. trial³ further indicated that surgical intervention may be more effective when received earlier in a patient that does not see results in the first 6 months of conservative treatment. In our case, if the patient’s symptoms did not improve or progressively deteriorated 6 months post-partum with conservative treatment, she would be referred back to her family physician for further assessment.

A limitation we identified in the RCT by Goeteyn et al.³ that would negatively influence the internal validity of the study, and the subsequent application of its findings to our patient, was the lack of control for baseline symptom duration. One of the inclusion criteria for participants was a minimum of 6 months of conservative therapy with no improvement, regardless of how long the patient had been suffering with TOS. Symptom duration ranged between 1 to 4.4 years among the majority of patients in the study, with a notable difference between treatment groups after randomization (see Appendix 1). For instance, the TA-TOD group had a median symptom duration of 1.5 years, whereas the conservative treatment group had a median symptom duration of 3 years. It can be postulated that the outcomes of surgical intervention were favored over conservative treatment in this study due to the greater chronicity of TOS symptoms in the conservative treatment group. Other limitations of this study were the lack of blinding and relatively small sample size,³ thereby further reducing the internal validity and generalizability of the results.

Evaluation of the Outcome

Based on the evidence reviewed,³ combined with our patient’s clinical state and circumstances, we would recommend that our patient receive conservative therapy for her TOS symptoms and be managed non-surgically for the duration of her pregnancy. She presented with predisposing factors of TOS such as pregnancy and stress (see Figure 2). It is possible that the patient’s symptoms would resolve with conservative therapy, in addition to giving birth and managing her stress. We would monitor her symptoms using functional outcome and pain disability questionnaires to track her progress through treatment. If her symptoms did not improve or deteriorated within six months post-partum with conservative therapy, the patient would be referred to her family physician for further evaluation and management.

Limitations

A limitation of our report is that our literature search used a 5-year inclusion criteria cut-off, which may have resulted in important articles being missed. Our search was also restricted to a single RCT, whereas a systematic review of RCTs would have been a higher level of evidence to help address our clinical question.^8,9,20 However, we repeated our literature search with no time cut-off and found no additional RCTs or systematic reviews of RCTs comparing surgical versus conservative treatment of TOS. Regardless, our literature search included only 1 database and did not consider alternative terms (e.g., ‘TOS’), or combinations of terms, that may have yielded additional relevant results. Moreover, use of the term ‘conservative therapy’ may not have captured articles of relevant interventions that were not explicitly classified by study authors as “conservative therapy.” Another limitation was that our report was based on a patient scenario rather than an actual patient. Furthermore, although upper extremity provocative maneuvers for TOS including the Adson’s and upper limb tension tests have been shown to have good diagnostic utility,^24,25 the clinical diagnosis of TOS remains controversial.^3,26,27 Corroboration of the patient’s clinical findings with further diagnostic testing (e.g., magnetic resonance imaging, Doppler ultrasonography, computed tomography angiography) would be needed to rule out other possible causes of her symptoms and confirm a diagnosis of true (rather than disputed²⁶) TOS. Furthermore, our proposed treatment plan was tailored to the unique clinical state and circumstances of an individual case and should not be extrapolated to the management of other TOS patients. Our report was also based on a patient suffering from symptoms related to the interscalene triangle and may not be applicable to other TOS presentations. Finally, because our report was based on the findings of one small RCT,³ clinicians should be wary that further research and discussion regarding the efficacy and effectiveness of conservative versus surgical treatment of TOS is needed. Our primary aim was to illustrate the process of incorporating research findings into clinical decision-making rather than provide a narrative literature review on the therapeutic management or diagnosis of TOS.²⁸ For further discussion on the etiology, pathogenesis, diagnosis, and surgical and non-surgical treatment of TOS, we refer readers to other publications.^{1,2,4–7,24–27}

CONCLUSION

Using an evidence-based format, the purpose of our article was to illustrate to clinicians how research findings can be applied to individual case management. In our case scenario, a 29-year-old pregnant woman presented with signs and symptoms and diagnostic findings consistent with TOS. By using the results of one recent RCT, in combination with our patient’s clinical state and circumstances, we deemed that conservative treatment would be appropriate for the duration of her pregnancy or until her symptoms improved. Progressive neurological deficits or lack of improvement six months post-partum would indicate a referral to her family physician for further evaluation and management.