A Prospective Evaluation of the Effect of Supervised Hand Therapy After Carpal Tunnel Surgery

Abstract

Background: The objective of this investigation is to examine the effect of postoperative therapy after routine carpal tunnel release. Our hypothesis was that supervised hand therapy does not improve outcomes after routine carpal tunnel release. Methods: Patients with carpal tunnel syndrome were randomly assigned to one of 3 groups based on the last digit of their medical record numbers to one of 3 groups: standard 6-week postoperative rehabilitation (standard therapy), expedited one-session postoperative rehabilitation group (expedited therapy), and no postoperative rehabilitation group (no therapy). The primary outcome measures were Quick Disabilities of the Arm, Shoulder, and Hand (QuickDASH) and return to work. The outcome questionnaire was completed preoperatively, at the 2-week follow-up visit, and monthly to 6 months after surgery. Results: All 3 treatment groups had similar mean QuickDASH scores preoperatively. At 1- to 6-month follow-up, all 3 groups had similar QuickDASH scores at each visit, and all showed a significant decline from baseline (preoperative) QuickDASH score. Overall, QuickDASH score decreased significantly from a preoperative visit mean of 42.7 to a final postoperative (visit 8) mean of 6.69. There was no significant difference in the mean QuickDASH score among all 3 groups at 6-month follow-up. There was no significance in the time of return to work among the 3 groups (standard therapy, 21.8 days; expedited therapy, 20.9 days; no therapy, 16.6 days). Conclusions: This investigation adds evidence that supervised hand therapy does not improve the outcomes of routine carpal tunnel surgery as measured by QuickDASH and return to work.

Introduction

According to the World Health Organization, the total expenditure on health care in the United States as a percentage of the gross domestic product was 17% in 2012.¹ This expenditure rate continues to rise as all stakeholders are working on identifying a sustainable solution to this cost crisis.^2-5 Porter and colleagues propose that emphasis should be placed on increasing the value of care delivered by reducing cost while improving outcomes.⁴ They propose transforming health care delivery by basing competition on value, which drives industry participants optimize the quality and efficiency of their service. To achieve this objective, physicians must justify their treatment algorithms with outcome measures, particularly if a process (eg, treatment algorithm) is modified to increase efficiency and decrease cost.

The effect of supervised therapy on the outcomes associated with carpal tunnel release is controversial.⁶ In 1993, Nathan et al reported that early postoperative therapy after carpal tunnel release is more critical than incision size in regaining normal function and decreasing the time loss cost.⁷ Since this report, multiple studies have investigated the benefit of several postoperative therapy modalities, including splinting, electrical stimulation, cold treatment, contrast baths, extremity elevation, topical application of herbal remedies, and a comprehensive rehabilitation regimen.^6-24 Some of these investigations suggest that postoperative rehabilitation is beneficial, while most of the studies conducted suggest that there is no benefit and that the additional cost is unjustifiable. In 2013, Peters et al performed a comprehensive review of the studies that have investigated the benefit of postoperative rehabilitation after carpal tunnel release.⁶ They reported that the evidence regarding postoperative interventions after carpal tunnel release is limited and of low quality, and therefore, the decision to implement a postoperative intervention should be based on the physician’s judgment and the patient’s preference. The objective of this investigation is to examine the effect of postoperative therapy after routine carpal tunnel release. Our hypothesis was that supervised hand therapy does not improve outcomes after routine carpal tunnel release.

Methods

After obtaining approval from our institution review board, 66 patients were randomly assigned to one of 3 groups based on the last digit of their medical record numbers (Table 1)—group 1: standard 6-week postoperative rehabilitation (standard therapy); group 2: expedited one-session postoperative rehabilitation group (expedited therapy); and group 3: no postoperative rehabilitation group (no therapy). Inclusion criteria included: (1) clinical evaluation consistent with carpal tunnel syndrome; and (2) electromyography (EMG)/nerve conduction study (NCV) classifying the severity of carpal tunnel syndrome. Exclusion criteria included: (1) history of cervical spine disorders or a positive Spurling test; (2) prior carpal tunnel release on ipsilateral hand; (3) signs of thenar atrophy on examination; and (4) workers’ compensation case.

Table 1.

Postoperative Hand Therapy Regimens.

Standard therapy	4-6 wk	Postoperative day 10 • Scar management: Micropore tape; scar massage • Gelflex brace for functional activities that involve pressure to palm (ie, grocery carts, driving) • Patient education: Appropriate activity level including instruction regarding overuse/underuse, wound care, and edema management • Differential tendon gliding exercises • Thumb and wrist range of motion • Median nerve gliding exercises • Orthosis at night and as needed during day for higher activity levels 2-6 weeks postoperatively • Initiate light resistive use and progress as tolerated. No repetitive gripping or heavy gripping • Patient education for proper ergonomics/body mechanics in preparation for return to work • Progress functional use (ie, activities of daily living as tolerated)
Expedited therapy	1 session	Postoperative day 10 • Scar management : Micropore tape (use tape until 2 months postoperatively); scar massage • Gelflex brace for functional activities that involve pressure to palm (ie, grocery carts, driving) • Patient education: Appropriate activity level including instruction regarding overuse/underuse, wound care, and edema management • Differential tendon gliding exercises • Thumb and wrist range of motion • Median nerve gliding exercises • Orthosis at night and as needed during day for higher activity levels • Review appropriate activity level to be sure of not overusing or underusing
No therapy	None	None

The primary outcome measures were Quick Disabilities of the Arm, Shoulder, and Hand (QuickDASH) and return to work. The outcomes questionnaire was filled out preoperatively, at the 2-week follow-up visit, and then monthly to 6 months after surgery. The questionnaire was distributed in person, through e-mail, or verbally over the phone (if subjects did not have an e-mail address). Return to work was self-reported by patients. Study data were collected and managed using the REDCap (Research Electronic Data Capture) program, which is a secure, web-based application designed to support data capture for research studies.

Surgical Technique

The carpal tunnel releases were all performed by the 2 senior authors through the same mini-open technique (1.5-2 cm).²⁵

Statistical Analysis

Patients with carpal tunnel syndrome underwent standard therapy, expedited therapy, or no therapy postoperatively. A patient-reported QuickDASH pain score was calculated at each visit. The mean and standard error QuickDASH score was calculated for each cohort at each visit. Preoperative QuickDASH scores were normally distributed, as determined by the Kolmogorov-Smirnov test (P = .088). The QuickDASH scores between cohorts were compared using a 1-way analysis of variance (ANOVA) for each visit date. In addition, a multivariable factorial ANOVA was used to compare the QuickDASH score between groups, adjusted for age ⩾65, hand dominance, sex, and EMG severity score >2. Time to return to work was compared between treatment groups using 1-way ANOVA. Given that there was a high crossover rate between treatment groups (34%), both intention-to-treat and as-treated analyses were performed. For a priori power analysis based on a minimal clinically important difference in QuickDASH of 12 and a variance of 15.3, 26 patients were required in each group to achieve 80% power.^26,27 Statistical analyses were performed using SAS statistical software v.9.4 (SAS Institute, Inc. Cary, NC). Statistical significance was set at P < .05.

Results

Of the 67 patients who were enrolled in the study and completed monthly postoperative questionnaires, 22 patients were randomized to standard therapy (8 men, 14 women; 53.9±11.9 years old), 19 to expedited therapy (4 men, 15 women; 56.5±13.0 years old), and 26 to no therapy (9 men, 17 women; 55.2±16.1 years old) (Table 2). Forty-four patients remained in the initial treatment group they were assigned, while 23 crossed over. Of the patients who crossed over, 13 patients in the standard therapy group crossed over to no therapy, 2 patients from expedited therapy crossed over to no therapy, and 8 patients from no therapy crossed over to standard therapy. All 67 had preoperative visit and initial postoperative visit. Sixty-five patients followed up at 1 month postoperatively, 63 at 2 months, 61 at 3 and 4 months, 59 at 5 months, and 55 (82%) at 6 months.

Table 2.

Demographics.

	Standard therapy	Expedited therapy	No therapy
Male	8	4	9
Female	14	15	17
Age	53.9	56.5	55.2
Symptom duration, mo	33.5	57.2	19.1
Employed	16	14	21
Retired	1	3	4
Unemployed	3	2	0
Hand dominance
Right	18	16	21
Left	3	3	3
Operative side
Right	10	11	15
Left	10	6	10
EMG severity
Mild	0	1	0
Moderate	11	9	15
Moderate-severe	8	5	10
Severe	2	4	1

Note. EMG = electromyography.

Intention-to-treat analysis revealed that all 3 treatment groups had similar mean QuickDASH scores preoperatively on 1-way ANOVA (P = .1571) (Figure 1, Table 3). At the first postoperative visit, prior to any evaluation by hand therapy in any group, 1-way ANOVA revealed a significant difference in QuickDASH score (P = .0097), revealing that the expedited therapy group had the highest QuickDASH score, with a mean of 44.76. The multivariable factorial ANOVA confirmed this relationship (P = .0165) after adjustment for patient demographics (Table 4). At 1-month follow-up, the standard therapy and no therapy groups showed a significant improvement in QuickDASH from baseline, with means of 30.9 and 28.0, respectively. At 1- to 6-month follow-up, all 3 groups had similar QuickDASH scores (P > .05), and all showed significant decline from baseline (preoperative) QuickDASH score. Overall, the QuickDASH score decreased significantly from a preoperative visit mean of 42.7 to a final postoperative (visit 8) mean of 6.69 (P < .0001). There was no significant difference in the mean QuickDASH score between all 3 groups (P = .6431) at 6-month follow-up. Sixteen patients (73%) enrolled to standard therapy were employed prior to undergoing a carpal tunnel release, 14 patients (74%) enrolled to expedited therapy were employed, and 21 patients enrolled to no therapy were employed. There was no significance in time to return to work between the 3 groups (group 1, 21.8 days; group 2, 20.9 days; group 3, 16.6 days) (P = .7814).

Figure 1.

Intention-to-treat analysis revealed that all 3 treatment groups had similar mean QuickDASH scores preoperatively on 1-way analysis of variance (P = .1571).

Note. QuickDASH = Quick Disabilities of the Arm, Shoulder, and Hand; F/u = follow-up.

Table 3.

Mean QuickDASH Score ± Standard Deviation, by Initial Treatment Group, and Univariable ANOVA P Value.

Visit No.	Initial treatment group			Overall mean	Pr > F (univariable ANOVA)
	Standard therapy	Expedited therapy	No therapy
Preoperatively	45.95 ± 19.23	46.04 ± 16.49	37.59 ± 15.48	42.68 ± 17.32	.1571
2-wk F/u	30.89 ± 15.54	44.76 ± 21.13	28.03 ± 18.19	33.71 ± 19.35	.0097
1-mo F/u	20.11 ± 12.9	28.47 ± 19.57	18.09 ± 12.03	21.75 ± 15.28	.0653
2-mo F/u	14.14 ± 13.94	19.02 ± 16.29	13.85 ± 11.24	15.49 ± 13.66	.4099
3-mo F/u	11.36 ± 16.49	11.08 ± 8.51	13 ± 10.81	11.94 ± 11.91	.8489
4-mo F/u	11.36 ± 16.17	5.78 ± 6.18	9.69 ± 12.02	8.94 ± 12	.354
5-mo F/u	8.92 ± 12.27	5.68 ± 5.9	8.24 ± 11.45	7.65 ± 10.27	.615
6-mo F/u	8.64 ± 13.4	6.55 ± 7.74	5.53 ± 8.69	6.69 ± 9.83	.6431

Note. QuickDASH = Quick Disabilities of the Arm, Shoulder, and Hand; ANOVA = analysis of variance; F/u = follow-up.

Table 4.

Multivariable Factorial ANOVA: Controlling for Sex, Hand Dominance, Age, and Severity (P Values Comparing Treatment Groups).

	Preoperatively	2-wk F/u	1-mo F/u	2-mo F/u	3-mo F/u	4-mo F/u	5-mo F/u	6-mo F/u
Sex	.3651	.1295	.0983	.8513	.6872	.6748	.5657	.9943
Age >65	.7771	.3369	.2317	.0738	.1572	.3244	.5338	.2297
EMG >2	.4123	.7337	.9613	.6469	.6506	.8958	.8575	.979
Dominant hand	.193	.5996	.4876	.4877	.6673	.6438	.8393	.9681
Initial treatment group	.2046	.0165	.1235	.509	.8416	.355	.6754	.5703

Note. ANOVA = analysis of variance; F/u = follow-up; EMG = electromyography.

Similarly, as-treated analysis revealed that all 3 treatment groups had similar mean QuickDASH scores preoperatively on 1-way ANOVA (P = .364) (Figure 2). At 1-month follow-up, the QuickDASH scores in the standard and no therapy groups (34.1 and 28.2, respectively) were again significantly lower than the QuickDASH score in the expedited therapy group (44.14). Otherwise, there was no significant difference between the 3 groups at 2- to 6-month follow-up (P > .05).

Figure 2.

As-treated analysis revealed that all 3 treatment groups had similar mean QuickDASH scores preoperatively on 1-way analysis of variance (P = .364).

Note. At 1-month follow-up, the QuickDASH scores in groups 1 and 3 (34.1 and 28.2, respectively) were significantly lower than the QuickDASH score in group 2 (44.14). Otherwise, there was no significant difference between groups at 2- to 6-month follow-up (P > .05). QuickDASH = Quick Disabilities of the Arm, Shoulder, and Hand; F/u = follow-up.

Discussion

The objective of this study was to examine the effect of postoperative therapy after routine carpal tunnel release. Our hypothesis was that supervised hand therapy does not improve outcomes after routine carpal tunnel release. At 1- to 6-month follow-up, all 3 groups had similar QuickDASH scores (P > .05), and all showed significant decline from baseline (preoperative) QuickDASH score. Overall, QuickDASH score decreased significantly from a preoperative visit mean of 42.7 to a final postoperative mean of 6.69 (P < .0001). There was no significant difference in the mean QuickDASH score between all 3 groups (P = .6431) at 6-month follow-up. In addition, there was no significant difference in time to return to work between the 3 groups (standard therapy, 21.8 days; expedited therapy, 20.9 days; no therapy, 16.6 days) (P = .7814). Based on our results, supervised hand therapy does not appear to improve the outcomes of routine carpal tunnel surgery.

In 1993, Nathan et al reported that early postoperative therapy is more critical than incision size in regaining normal function and decreasing the time loss cost.⁷ They examined the outcomes of carpal tunnel release performed through either short (<2.5 cm) incision or longer (>2.5 cm) incision, both combined with a supervised postoperative rehabilitation protocol. They reported that early hand therapy was more important than a short incision for recovering hand function and minimizing time loss cost. In contrast to these results, our data indicate that supervised hand therapy does not improve QuickDASH score and does not facilitate earlier return to work.

In 2007, Pomerance et al conducted a randomized trial to compare the difference in outcomes of carpal tunnel surgery with or without supervised hand therapy.¹⁴ They performed the surgery through a short incision. The patients assigned to the supervised hand therapy group were prescribed 6 sessions of therapy over a 6-week course beginning 5 to 7 days after surgery. They reported no difference in DASH score, pain score, and grip and pinch strength at 2-week, 4-week, 6-week, 3-month, and 6-month follow-up. These conclusions compare favorably with our results, which indicate that supervised hand therapy does not improve the outcomes of routine carpal tunnel surgery.

There are several limitations of this investigation. The primary limitation of this investigation is the 34% (n = 23) crossover rate between treatment groups. This introduces bias to the groups by leading patients who are doing worse to seek therapy and patients who are doing well to cross over to the no therapy group. However, both intention-to-treat and as-treated analyses did not reveal significant differences between groups at 2- to 6-month follow-up. Second, given the small size of each group, we did not stratify employment based on labor intensity. Third, we chose to use QuickDASH and return to work as our primary outcome measures. We acknowledge that a carpal tunnel–specific questionnaire could have been more sensitive at detecting differences between the treatment arms in our study. However, choosing QuickDASH allowed us to compare our results with previous publications that examined the effect of postoperative rehabilitation after carpal tunnel surgery. Furthermore, while we could have included these instruments, previous publications have reported that pitfalls of studies depend on questionnaires that patients perceive as overly burdensome.²⁸ Fourth, our a priori power analysis required an enrollment of 26 patients in each group. During our enrollment period, 22 patients were enrolled to standard therapy, 19 to expedited therapy, and 26 patients to no therapy. Therefore, our study was underpowered and must be interpreted with this limitation in mind.

This study adds evidence that supervised hand therapy does not improve the outcomes of routine carpal tunnel surgery as measured by QuickDASH and return to work. As hand surgeons consider streamlining the treatment algorithm for carpal tunnel syndrome, supervised hand therapy should be not be routinely prescribed to all patients after carpal tunnel release.