Relationship between pain,nerve injury and clinical outcomes after flexor tendon injuries in zones 1-2: a retrospective cohort study

Abstract

Introduction

The influence of pain and a concomitant digital nerve injury on the course of rehabilitation after flexor tendon injury remains ambiguous. The objectives of the study were to: i) analyse the evolution of pain spanning one year after a primary flexor tendon repair in zones 1–3; ii) examine the differences in pain levels in patients with and without digital nerve injury; and iii) evaluate the relationship between pain, digital nerve injury and pain medication, total active motion (TAM), DASH scores and patient satisfaction.

Methods

Data from 189 patients were retrieved from a flexor tendon registry between 2014 and 2020. Differences in pain, TAM, DASH and patient satisfaction were analysed. Multiple linear and binary logistic regression analyses were performed to determine the relationship between clinical outcomes.

Results

Pain significantly decreased in the course of rehabilitation (p < 0.001 to 0.006). No relationship could be identified between nerve injury and pain (p = 0.21–0.97). In week 6, the presence of pain and a nerve injury were significantly associated with lower TAM scores (p = 0.001). In week 13, pain during motion (p < 0.001) and the presence of a nerve injury (p = 0.036) were significantly associated with worse DASH scores. Patient satisfaction was significantly inversely correlated to pain during motion in weeks 13 and 26 (p < 0.001).

Conclusion

We found a significant relationship between pain during motion and pain medication intake, TAM, DASH scores and patient satisfaction. It is therefore advisable to closely monitor this parameter after flexor tendon injuries.

Study registration

This multi-center cohort study is registered under https://clinicaltrials.gov: NCT04312412.

Keywords

flexor tendon repair rehabilitation pain nerve injury

Introduction

Flexor tendon injuries of the hand are relatively common but demanding injuries for both the patient and the clinical hand therapist.^1,2 From the patient’s perspective, former experiences, limited hand use, and anxiety linked to concerns regarding permanent functional loss and pain may impact recovery after injury.³ The therapists’ expertise during the different rehabilitation regimens is fundamental to safely guiding the patients through the recovery process, in conjunction with a continuous and standardized measurement of the treatment outcomes.^4–6 In this day and age, patient-reported outcome measures (PROMs), such as satisfaction with treatment results and pain should be documented in support of clinician-reported outcome measures (CROMs), such as grip and pinch strength or range of motion.⁷ The patient’s perceived level of pain is an important factor that guides the therapeutic process and may delay or even hinder the recovery of finger motion and resumption of daily activities. Pain is defined as “an unpleasant sensory and emotional experience associated with, or resembling that associated with, actual or potential tissue damage,” (p 1976).⁸ However, pain after surgery is described as a sequela to an inflammatory process and isolated injury to nerves.⁹ It can be expected to be controlled and resolved within one week post-surgery.¹⁰ To better understand the course of pain after flexor tendon surgery, it is imperative for a hand therapist to routinely document and address both pain at rest and pain caused by movement during rehabilitation. Pain triggered by movement, for example, could be caused by tissue damage, potentially getting better over time, while pain at rest might be caused by nerve laceration or tissue inflammation. Therefore, pain may potentially hinder active range of motion (aROM) as well as generating an emotional burden for the patient.

Concomitant digital nerve injury occurs in 25%–43% of flexor tendon lacerations in zones 1–3.^1,11–14 Previous studies on the outcome after flexor tendon repair reported contradictory results on the association between concomitant nerve lacerations and different CROMs and PROMs.^12–15 For example, Trumble et al.¹² stated that a concomitant nerve injury resulted in reduced aROM and worse patient satisfaction, while Rigo and Røkkum,¹⁵ Rosberg et al.¹¹ and Keane et al.¹⁶ concluded that a nerve injury had no influence on the functional recovery. However, it remains unclear if a concomitant digital nerve injury affects pain. Although there is an emphasis on addressing pain during the course of rehabilitation,¹⁷ the distinction between pain at rest and during motion has often been neglected in the past. Therefore, the primary purpose of this study was to document how pain at rest and during motion changes over the course of recovery after primary flexor tendon repairs in zones 1–3. The secondary objective was to explore if there is a difference in the level of pain at rest and during motion in patients with and without a concomitant digital nerve injury in weeks 1, 2, 3, 6, 13, 26, and 52. The third objective was to evaluate if pain medication intake, total active motion (TAM), self-defined functional recovery and patient satisfaction were affected by the level of pain or by the presence of a concomitant digital nerve injury during the first year after surgery.

Methods

The participants of this retrospective cohort study represent a subgroup of patients from a Swiss flexor tendon registry who were treated at three public university hospitals since 2014. Inclusion criteria were: (i) signed informed consent, (ii) aged between 18 and 75 years, (iii) primary finger flexor tendon injury in zones 1–3, (iv) ≥50% of the flexor digitorum profundus (FDP) tendon injured, treated (v) with an immediate active motion regime (CAM) using a long dorsal dorsal-blocking splint with the wrist in 20° of extension, and the MCPs in 40° of flexion, (vi) in one of the three hand therapy units of the participating hospitals. Data were analyzed from January 2014 to December 2020 and only if ≥80% of data relating to pain and pain medication were present. Therapy started within 2–5 days after surgery. All patients were prescribed oral pain medication (paracetamol and metamizole) and instructed by their hand surgeon to take them in a standardized manner. Concomitant injuries to digital nerves that underwent surgery did not alter the therapy protocol. Exclusion criteria were the presence of (i) concomitant fractures (e.g. avulsion fracture of FDP), (ii) laceration of FPL, (iii) diagnosis of rheumatological conditions, (iv) more than one finger involved and (v) finger replantation.

Data collection

Clinical data were routinely collected from week 1 up to one-year post-surgery during hand therapy visits in one of the three clinics by occupational and physical therapists specialized in hand therapy. Pain (at rest and during motion) was collected in weeks 1, 2, 3, 6, 13, 26 and 52; aROM in weeks 6, 13, 26 and 52; patient satisfaction, self-reported physical functioning, and symptoms in weeks 13, 26 and 52.

Patient-reported outcome measures

Verbal Numeric Rating Scale (VNRS)

In our hand therapy units, we use the Verbal Numeric Rating Scale as a more practical alternative for assessing pain at rest and during motion in everyday clinical practice instead of the ‘gold standard’ Visual Analogue Scale (VAS). It is rapid, requires no instruments and does not require intact motor skills. The VNRS has demonstrated that it correlates well with the VAS (r = 0.95, 95% CI 0.94–0.96) and it is as sensitive for detecting clinically important changes in pain.¹⁸ The minimum clinically important difference (MCID) in VNRS for acute pain is 1.4 points (95% CI 1.2–1.6),¹⁸ being slightly lower than the established MCID in a non-shoulder hand and upper extremity postoperative population after elective hand surgery (range between 1.6 and 1.9 points on the VAS).¹⁹ Patients were asked to score their pain on the VNRS from zero to 10 with zero indicating “no pain” and 10 indicating “worst pain imaginable.” The scores are interpreted as: “no pain” (0), “mild pain” (1–3), “moderate pain” (4–6), “severe pain” (7–10).^20,21 In addition, patients were asked each week if they had taken pain medication or not (“yes” or “no” questions without further indication of the kind and dosage of medication).

Patient satisfaction

We asked our patients to respond to the following question: “Today, how satisfied are you with your hand in general?” on an 11-point Likert scale ranging from 0 to 10 (10 indicating maximum satisfaction).

Self-reported questionnaire regarding physical function and symptoms (DASH)

The Disabilities of the Arm, Shoulder and Hand Outcome Measure is a 30-item, self-reported questionnaire designed to measure physical functioning and symptoms in patients with single or multiple musculoskeletal disorders of the upper limb.²² Scores range from 0 (no disability) to 100 (completely disabled). The MCID of the DASH after hand surgical interventions ranges from 10.8 to 18.8 points.¹⁹

Clinician-reported outcome measures

Active range of motion (aROM) and ability to reach the palm of the hand

Measurement of finger joint mobility was assessed using manual goniometry. The total active motion (TAM) was calculated as the sum of degrees patients can actively flex and extend their proximal and distal interphalangeal joints. Then, the TAM scores were transferred to the proposed classification system by Tang,¹⁷ where the return of motion is expressed in percentages to the contralateral hand: 1 = “excellent” (90–100%), 2 = “good” (70–89%), 3 = “fair” (50–69%), 4 = “poor” (30–49%) and 5 = “failure” (<30%). Fingertip to palm mobility of the injured finger, where a ‘touch-down’ was defined as the affected fingertip touching the palm, was noted as “yes” or “no” by the therapist.

Data analysis

Demographics were analyzed using descriptive statistics. Means and standard deviations (SD), medians and interquartile ranges (IQR) for continuous variables and percentages for categorical variables were computed. As the Shapiro-Wilk test did not show a normal distribution of pain at rest and during motion, the non-parametric Friedman test was applied to calculate differences between all seven measurement time points for pain at rest and during motion, followed by post-hoc tests (Wilcoxon-sign-rank-test with Bonferroni adjustment). Small, medium and large effect sizes (ES) were calculated as d = 0.20, 0.50, and 0.80.²³ Differences in the level of pain at rest and during motion in patients with and without a concomitant digital nerve injury were evaluated for each of the seven measurement time points using the Mann-Whitney U-test. Blockwise multiple linear and binary logistic regression models were performed to determine the relationship between pain at rest and during motion (block 1) and the presence of a nerve injury (block 2) on pain medication intake, TAM scores, touch-down, DASH and satisfaction scores. One model was calculated per measurement time point, using the corresponding outcomes from that time point (e.g. level of pain at week 6 and TAM scores at week 6). Statistical significance was set at p ≤ 0.05. The guidelines from Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) Statement were used for this study.²⁴

Results

Of the 308 participants in the data pool, 189 patients were included in this study and further divided into two subgroups: with and without concomitant digital nerve laceration (Figure 1). 72.5% of injuries occurred in zone 2 and 27.5% in zone 1. All zone 3 injuries in the data pool (n = 38) did not fulfil the inclusion criteria. Demographics, injury and surgery characteristics are presented in Table 1.

Figure 1.

Study flow chart.

Table 1.

Demographics, injury and surgery characteristics.

Characteristics	All patients	Patients with nerve laceration	Patients without nerve involvement
Characteristics	n = 189 (189 fingers)	n = 95 (95 fingers)	n = 94 (94 fingers)
Age in years, mean (SD) [range]	37.0 (4.0)	38.3 (14.6)	35.8 (13.4)
Age in years, mean (SD) [range]	[18–74]	[18–74]	[18–73]
Gender, n males (%)	126 (66.7)	66 (69.5)	60 (63.8)
Blue collar, n (%)	113 (62.4)	61 (66.3)	52 (58.4)
Dominant hand injured, n (%)	96 (52.7)	42 (46.1)	54 (59.3)
Digital nerve injured, n (%)	95 (50.3)	95 (100)	0 (0)
Involved digit, n (%)
Dig II (zones 1 - 2)	65 (34.4) (16–49)	40 (42.1) (11–29)	25 (26.6) (5–20)
Dig III (zones 1–2)	26 (13.7) (8–18)	10 (10.5) (5–5)	16 (17.0) (3–13)
Dig IV (zones 1–2)	30 (15.9) (14–16)	21 (22.1) (11–10)	9 (9.6) (3–6)
Dig V (zones 1 - 2)	68 (36.0) (14 – 54)	24 (25.3) (6–18)	44 (46.8) (8–36)
Zone 1 injury, n (%)	52 (27.5)	33 (34.7)	19 (20.2)
Zone 2 injury, n (%)	137 (72.5)	62 (65.3)	75 (79.8)
FDP, number of strands
2-Strand, n (%)	2 (0.1)	1 (0.1)	1 (0.1)
4-Strand, n (%)	35 (18.5)	16 (16.8)	19 (20.2)
6-Strand, n (%)	124 (65.6)	61 (64.2)	63 (67.0)
8-Strand, n (%)	3 (1.6)	2 (0.2)	1 (0.1)
10-Strand, n (%)	2 (0.1)	1 (0.1)	1 (0.1)
Pull-out, n (%)	20 (10.6)	12 (12.6)	8 (8.5)
Mitek bone anchor, n (%)	3 (0.2)	2 (0.2)	1 (0.1)
Mechanism of injury, n (%)
Clean cut	162 (85.7)	82 (86.3)	80 (85.1)
Mild crush	18 (9.5)	7 (7.4)	11 (11.7)
Crush	9 (4.8)	6 (6.3)	3 (3.2)
Anesthesia
Plexus, n (%)	98 (52.4)	49 (52.1)	49 (52.7)
General, n (%)	40 (21.4)	22 (23.4)	18 (19.3)
Local, n (%)	49 (26.2)	23 (24.5)	26 (28.0)
Complications
Rupture, n (%)	19 (10.0)	6 (6.3)	13 (13.8)
Tenolysis, n (%)	13 (6.9)	5 (5.3)	8 (8.5)

SD = standard deviation, n = number, bold = most frequent occurrence of characteristic.

Change of pain at rest and pain during motion over the course of rehabilitation

Results of the Friedman test showed differences between all seven measurement time points for both pain at rest (Chi-Square₍₆₎ = 222.2, p ≤ 0.001, n = 129) and during motion (Chi-Square₍₆₎ = 201.1, p ≤ 0.001, n = 82). Both variables continuously decreased over the course of rehabilitation, with the median pain at rest from 1 to 0 and the median pain during motion from 3 to 0 (Table 2). Post-hoc tests showed significant differences from weeks 1-2 and 2-3 for both pain at rest and during motion and additionally for pain during motion between weeks 3 and 6 with small effect sizes (Table 3). Categorizing the amount of pain according to Sloman et al.²⁰ most patients perceived mild pain at rest in the first three weeks post-surgery and were all pain-free thereafter. Pain during motion constantly decreased from moderate (first two weeks in some patients) to mild (up to 26 weeks in most patients) to no pain (after 26 weeks for all patients) (Table 2).

Table 2.

Median and mean pain scores at rest and while moving in all patients and subgroups with differences between subgroups (Mann-Whitney-U-Scores).

Measurement time point	All patients				Pain at rest					Pain during motion
	Pain at rest		Movement evoked pain		With nerve laceration	Without nerve involvement	Mann-Whitney-U-Test			With nerve laceration	Without nerve involvement	Mann-Whitney-U-Test
	n	Median (IQR)	n	Median (IQR)	Median (IQR)	Median (IQR)	n	p-value	ES (d)	Median (IQR)	Median (IQR)	n	p-value	ES (d)
	n	Mean [SD]	n	Mean [SD]	Mean [SD]	Mean [SD]	n	p-value	ES (d)	Mean [SD]	Mean [SD]	n	p-value	ES (d)
Week 1	145	1 (0–2)	143	3 (1–5)	1 (0–2)	1 (0–2)	145	0.951	0.04	3 (1–5)	3 (1–5)	143	0.774	0.02
Week 1	145	1.5 [2.0]	143	3.24 [2.53]	1.54 [2.10]	1.46 [1.93]	145	0.951	0.04	3.22 [2.65]	3.27 [2.24]	143	0.774	0.02
Week 2	158	0 (0–1)	158	2 (0–4)	0 (0–2)	0 (0–1)	158	0.684	0.12	2 (0–4)	2 (0–4)	158	0.746	0.08
Week 2	158	0.91 [1.63]	158	2.56 [2.4]	1.00 [1.67]	0.81 [1.57]	158	0.684	0.12	2.47 [2.30]	2.66 [2.52]	158	0.746	0.08
Week 3	160	0 (0–1)	153	2 (0–3.5)	0 (0–0)	0 (0–1)	160	0.614	0.07	2 (0–4)	2 (0–3.25)	153	0.97	0.00
Week 3	160	0.64 [1.38]	153	2.07 [2.15]	0.59 [1.30]	0.68 [1.46]	160	0.614	0.07	2.07 [2.18]	2.07 [2.14]	153	0.97	0.00
Week 6	155	0 (0–0)	149	0 (0–3)	0 (0–0)	0 (0–0)	155	0.746	0.03	0 (0–2.75)	1 (0–3)	149	0.209	0.13
Week 6	155	0.33 [1.0]	149	1.49 [1.94]	0.34 [1.08]	0.31 [0.92]	155	0.746	0.03	1.37 [1.99]	1.62 [1.91]	149	0.209	0.13
Week 13	150	0 (0–0)	125	0 (0–1)	0 (0–0)	0 (0–0)	150	0.225	0.25	0 (0–1.25)	0 (0–1)	125	0.281	0.31
Week 13	150	0.11 [0.62]	125	0.86 [1.72]	0.18 [0.82]	0.03 [0.17]	150	0.225	0.25	1.11 [2.02]	0.59 [1.26]	125	0.281	0.31
Week 26	150	0 (0–0)	115	0 (0–0)	0 (0–0)	0 (0–0)	150	0.244	0.23	0 (0–0)	0 (0–0)	115	0.647	0.16
Week 26	150	0.07 [0.46]	115	0.45 [1.35]	0.11 [0.61]	0.01 [0.12]	150	0.244	0.23	0.55 [1.41]	0.34 [1.28]	115	0.647	0.16
Week 52	148	0 (0–0)	110	0 (0–0)	0 (0–0)	0 (0–0)	148	0.875	0.05	0 (0–0)	0 (0–0)	110	0.666	0.02
Week 52	148	0.05 [0.38]	110	0.15 [0.72]	0.06 [0.46]	0.04 [0.27]	148	0.875	0.05	0.16 [0.85]	0.15 [0.56]	110	0.666	0.02

n = number, SD = standard deviation, IQR = interquartile ranges, ES = effect size, where small, medium and large are d = 0.20, 0.50, and 0.80, level of significance = p ≤ 0.05.

Pain categorization by Sloman et al.²⁰: “no pain” (0), “mild pain” (1–3), “moderate pain” (4–6), “severe pain” (7–10).

Table 3.

Differences between pain at rest and pain during motion over time of all patients (Wilcoxon sign rank test scores).

Paired outcome measurements		Pain at rest				Pain during motion
Paired outcome measurements		n	Z score	p-value sig. (2-Tailed)	ES (d)	n	Z score	p-value sig. (2-Tailed)	ES (d)
Pair 1	week 2–week 1	129	−4.35	<0.001*	0.31	82	−4.40	<0.001*	0.28
Pair 2	week 3–week 2	129	−2.73	0.006*	0.18	82	−2.71	0.007*	0.22
Pair 3	week 6–week 3	129	−2.25	0.024	0.26	82	−3.71	<0.001*	0.28
Pair 4	week 13–week 6	129	−1.60	0.110	0.26	82	−1.76	0.078	0.34
Pair 5	week 26–week 13	129	−0.81	0.417	0.07	82	−1.96	0.051	0.27
Pair 6	week 52–week 26	129	0.00	1.00	0.05	82	−1.66	0.096	0.28

n = number, ES = effect size, where small, medium and large are d = 0.20, 0.50, and 0.80.

Statistically significant changes over time adjusted for multiple comparison (p ≤ 0.008).

Differences in level of pain at rest and pain during motion in patients with and without a concomitant digital nerve injury

There were no statistically significant differences in pain at rest and during motion between patients with and without a concomitant nerve injury at any measurement time point. Effect sizes were small (pain at rest: p =0.225 to 0.951, ES d = 0.03 to 0.25; pain during motion: p =0.209 to 0.970, ES d = 0.00 to 0.31) (Table 2).

Relationship between concomitant nerve injury, pain level and pain medication intake

In the first week post-surgery, 72.5% of patients with (n = 68) and 70.4% without (n = 66) nerve involvement took pain medication. In week 2, 37.5% (n = 35) with and 29.5% (n = 27) without concomitant nerve injury were still using pain medication, decreasing to 28.3% (n = 27) and 22.7% (n = 21) in week 3 and to 12.5% (n = 11) and 11.4% (n = 10) in week 6, respectively. From week 13 onwards, only one patient (1.8%) with nerve injury continued taking pain medication for up to one year, while no patient without nerve involvement was still on pain medication during this time span.

The binary logistic regression model showed that experiencing pain during motion in weeks 2, 3 and 6 and pain at rest in week 2 was significantly associated with pain medication intake (Table 4). For example, pain medication intake increased by 34.6% (95% CI [1.04, 1.74]) in week 2 for a one-unit increase in pain at rest. For a one-unit increase in movement-evoked pain, pain medication intake steadily increased over time by 29.1% (95% CI [1.08, 1.55]) in week 2, 84.4% (95% CI [1.43, 2.38]) in week 3 and 117.5% (95% CI [1.52, 3.12]) in week 6. The likelihood of taking pain medication was always more likely with a concomitant nerve injury than without, being 65 times more likely at week 1, 76-fold at week 2, 49-fold at week 3 and 36-fold at week 6. For the subsequent measurement time points (weeks 13, 26 and 52), a regression model could not be calculated, as only one person continued taking pain medication.

Table 4.

Binary logistic regression models for the outcome measurements “pain medication and ‘touch down.”

	Time points	Model parameters				Pain at rest			Pain during motion			Nerve injury
	Time points	n	χ ² ₍₃₎	p-value	f ²	OR	95% CI	p-value	OR	95% CI	p-value	OR	95% CI	p-value
Pain medication	week 1	131	13.73	0.003*	0.16	1.38	0.96, 2.01	0.086	1.17	0.95, 1.45	0.172	1.53	0.68, 3.46	0.309
	week 2	147	26.21	<0.001*	0.29	1.35	1.04, 1.74	0.024*	1.29	1.08, 1.55	0.006*	1.32	0.62, 2.82	0.479
	week 3	141	37.5	<0.001*	0.55	0.98	0.69, 1.39	0.913	1.84	1.43, 2.38	<0.001*	2.06	0.79, 5.35	0.139
	week 6	140	30.04	<0.001*	0.64	0.80	0.49, 1.32	0.385	2.18	1.52, 3.11	<0.001*	2.81	0.74, 10.73	0.131
Touch down	week 6	123	13.06	0.001*	0.16	0.82	0.49, 1.37	0.438	0.71	0.55, 0.92	0.009*	****
	week 13	104	8.31	0.016*	0.12	0.74	0.35, 1.57	0.433	0.76	0.60, 0.98	0.035*
	week 26	85**	9.58	0.002*	0.25				0.59	0.40, 0.86	0.007*
	week 52	***

N = number, χ² = Chi-Square Statistic, f² = effect size, OR = odds ratio, CI = confidence interval, f² = 0.02 (small), = 0.15 (medium), = 0.35 (strong) effect size.

Statistically significant (p ≤ 0.05). A regression model per time point was calculated as follows: e.g. pain at rest week 3+ pain during motion week 3 nerve injury as predictors of week 3 pain medication intake.

Only calculated with pain during motion (χ² (1)), as pain at rest did never show a statistically significant influence on “touch down.”

***

Week 52 was not calculable for “touch down” as all patients were able to perform one at this time point.

****

The binary logistic regression model for “touch down” with the independent variable “nerve injury” is not statistically significant: χ² (1) = 0.309, p =0.578, n = 189. Therefore, no further calculations have been conducted.

Relationship between concomitant nerve injury, pain level and “touch down”

We found no relationship between the presence of a nerve injury and the performance of a “touch-down” at any measurement time point (Table 3). Pain during motion had a significant association with the ability to perform a “touch down” (p = 0.009 week 6; p = 0.035 week 13; p = 0.007 week 26), but not pain at rest (p = 0.438 week 6, p = 0.433 week 13). For week 26, only 3.3% of all patients had pain at rest, therefore, the model was only calculated with pain during motion. Week 52 was not modelled as all patients were able to perform a ‘touch-down’ at this time point. For a one-unit increase in pain during motion, the relative probability of performing a “touch-down” decreased by 18.5% (95% CI [0.49, 1.37]) at week 6, 26.1% (95% CI [0.35, 1.57]) at week 13 and 41.5% (95% CI [0.4, 0.86]) after 26 weeks.

Relationship between concomitant nerve injury, pain level and TAM scores

The mean TAM scores in week 6 were 118.5 (SD 52.8) degrees, increasing to 139.2 (SD 51.7) at week 13, and 158.3 (SD 56.8) at week 26, reaching 189.0 (SD 54.7) degrees after one year. Interpreting TAM scores as the return of motion in percentage to the contralateral hand, our results showed that 58.9% of patients (n = 93) moved their finger “poor-to-fair” after six weeks, whereas in week 52, 93.3% of patients (n = 42) scored “good-to-excellent” on the Tang Score (Table 5). The presence of pain at rest and during motion with the presence of a nerve injury was significantly associated with lower TAM scores at 6 weeks post-surgery (Table 6). It was found that the TAM scores decreased on average 6.6 (SE 2.22) degrees for a one-unit increase in pain during motion. If no concomitant nerve injury was present, TAM scores increased on average by 17.3 (SE 8.64) degrees. However, only 9.4% of the variance in TAM scores can be explained by those three variables (pain at rest, pain during motion and the presence of a nerve injury), which corresponds to a small effect size of f² = 0.10. For all other measurement time points (weeks 13, 26 and 52), no statistically significant association between pain, the presence of a nerve injury and achieved TAM scores were found (p = 0.117 to 0.525). This was because only 5.3% of patients suffered from pain at rest at week 13, decreasing to 3.3% at week 26 and 2.1% after one year, while 90% of patients had no to mild pain during motion at week 13, increasing to 95.7% at week 26 and 99.1% after one year.

Table 5.

Tang criteria for assessment of total active movement (TAM).

Return of motion in percentage	week 6		week 13		week 26		week 52
Return of motion in percentage	n	%	n	%	n	%	n	%
Excellent [90–100% return of motion]	10	6.3	26	17.2	27	26.7	20	44.4
Good [70–89% return of motion]	38	24.0	51	33.8	44	43.6	22	48.9
Fair [50–69% return of motion]	52	32.9	51	33.8	23	22.8	3	6.7
Poor [30–49% return of motion]	41	26.0	16	10.6	6	5.9	—	—
Failure [<30% return of motion]	17	10.8	7	4.6	1	1.0	—	—

n = number of fingers, % = percentage.

The TAM scores are calculated as the sum of active ranges of motion of the PIP and DIP joints. % return of motion is the comparison to the contralateral side (or 175° if missing).

Table 6.

Multiple regression models for the outcome measurements TAM (total active motion), DASH and patient satisfaction scores.

	Time points	Model parameters						Pain at rest		Pain during motion		Nerve injury
	Time points	n	F₍₃₎	p-value	f ²	R-Square	Adjusted R-square	Beta	p-value	Beta	p-value	Beta	p-value
TAM scores	week 6	138	5.71	0.001*	0.10	0.113	0.094	−0.17	0.069	−0.17	0.065	0.17	0.047
DASH	week 13	101	7.80	<0.001*	0.19	0.173	0.157	0.16	0.116	0.35	<0.001*	0.20	0.036*
	week 26	68	6.78	0.002*	0.17	0.173	0.147	—	—	0.41	0.001*	0.08	0.457
	week 52	40	50.44	<0.001*	2.53	0.732	0.717	—	—	0.85	<0.001*	—	—
Satisfaction	week 13	107	11.2	<0.001*	0.29	0.246	0.224	0.12	0.228	0.41	<0.001*	0.08	0.354
	week 26	74	5.97	0.001*	0.20	0.204	0.169	0.37	0.012*	0.55	<0.001*	0.21	0.053
	week 52	38	3.74	0.020*	0.22	0.248	0.182	0.40	0.208	−0.69	0.036*	0.33	0.033*

TAM = total active motion, being the sum of PIP and DIP joint motion, n = number, F = F-statistic, f² = effect size, where f² = 0.02 (small), = 0.15 (medium), = 0.35 (strong) effect size.

Statistically significant (p ≤0.05). A regression model per time point was calculated as follows: e.g. pain at rest week 6 + pain during motion week 6 + nerve injury as predictors of week 6 TAM scores.

Relationship between nerve injury, pain and the DASH

The mean DASH scores showed low levels of disability already at week 13 (mean 17.0 ± 14.7 points), decreasing constantly from week 26 (mean 7.1 ± 7.8 points) to week 52 (mean 5.1 ± 8.6 points). For week 13, pain during motion and the presence of a nerve injury were statistically significantly associated with worse DASH scores (Table 6). If pain during motion increased by one point, the DASH scores increased on average by 2.1 points (SE 0.77). If no concomitant nerve injury is present, the DASH scores decreased on average by 4.6 points (SE 2.37) at week 13.15.7% of the variance in the DASH scores can be explained by this model, corresponding to a medium effect size. For those patients still experiencing pain during motion at weeks 26 and 52, DASH scores were on average slightly higher than for those patients without pain during motion, namely 2.1 points (SE 0.57) at week 26 and 5.8 points (SE 0.58) after one year per each one-unit increase in pain during motion.

Relationship between concomitant nerve injury, pain level and patient satisfaction

Patient satisfaction in all patients increased from 7.2 (SD 1.9) points in week 13, 8.1 (SD 1.6) points in week 26 up to 8.7 (SD 1.1) points in week 52. The multiple regression model showed a relationship between pain during motion, pain at rest, the presence of a nerve injury and patient satisfaction at weeks 13, 26 and 52 with medium effect sizes (Table 6). Pain during motion had a statistically significant effect on satisfaction at week 13 (p ≤ 0.001), causing an increase in patient satisfaction by every 0.5 points (SE 0.1) decrease in pain during motion. After 26 weeks, both pain during motion and pain at rest had a statistically significant effect on satisfaction, but not the presence of a nerve injury. If both pain during motion and at rest decreased by one point, patient satisfaction increased on average by 0.5 points (SE 0.14) and 0.9 points (SE 0.36), respectively. At week 52, patient satisfaction increased on average by 0.6 points (SE 0.29) per each point decrease in pain during motion.

Discussion

The primary purpose of this study was to analyse the evolution of pain at rest and during motion spanning one year after a primary flexor tendon repair in zones 1–3. As all zone 3 patients did not fulfil our inclusion criteria, our findings only apply to zone 1-2 injuries. Both pain at rest and during motion steadily decreased over time, with significant decreases up to 6 weeks post-surgery. Most patients perceived no to mild pain at rest during the first three weeks and were pain-free thereafter. Pain during motion was markedly higher than pain at rest in the first six weeks. Hah et al.²⁵ identified a pain level of ≥3 on day 10 after surgery as a cut-off score for predicting chronic post-surgical pain at day 90. In our cohort, patients reported median levels of pain at rest of one point in week 1, decreasing to zero thereafter. Median levels of pain during motion were three points in week 1, going down to two points in weeks 3 and 4 and to zero thereafter. Therefore, most of our patients were never at risk of developing chronic post-surgical pain.

The secondary objective was to examine the differences in pain at rest and during motion in patients with and without digital nerve injury. In our study, 50.3% of injured fingers had an additional concomitant digital nerve lesion. This rate is comparable with previous studies (range from 34.4 to 56.1%).^{1,13,15,26–30} We could not find a statistically significant association between a concomitant digital nerve injury and the course of pain over one year after primary finger flexor tendon repair in zones 1-2. Our study shows that pain at rest after flexor tendon repair is generally low in both patient groups with and without concomitant nerve injury. Pain was reported in previous studies in patients after flexor tendon surgery, but rarely distinguished between pain at rest and pain during motion.^26,31,32 Furthermore, these authors did not discriminate between patients with and without nerve involvement. Our study demonstrates that there is no need to make this distinction, as there were no statistically significant differences in perceived pain levels between patients with and without a concomitant nerve injury at any measurement time point. This result, however, might be influenced by the (on average) 56.5-fold higher chance of pain medication intake over the first 6 weeks post-surgery when a concomitant nerve injury was present. Sade et al.³¹ for example, reported mean VAS pain levels of 6.5 (SD 2.5) points at the beginning of therapy and 1.3 (SD 2.0) points 12 weeks after flexor tendon surgery. The average pain level was markedly higher than in our study, but it was not stated if this was pain at rest or during motion or if patients were on pain medication.

The final objective was to evaluate the relationship between pain at rest, pain during motion, concomitant digital nerve injury and pain medication, total active motion (TAM), self-defined functional recovery (DASH) and patient satisfaction. Our results showed that during the first two weeks post-surgery, most patients took pain medication. After 3 weeks, we observed a change in that pain at rest was also tolerated without pain medication. Although the likelihood for pain medication intake increased the longer the pain during motion lasted, a combination of oral pain medications (paracetamol and metamizole) seemed nevertheless to be efficient and sufficient for postoperative pain control. This is in line with the algorithm for hand surgery patients with acute postoperative pain developed by Kelley et al.³³ recommending oral pain medication for patients after soft-tissue surgery and no major pain history.

In our study, we found a significant relationship between the pain level, presence of a concomitant nerve injury and TAM at week 6 only, as most patients did not suffer from pain at rest after that time point. Additionally, the majority of patients had “no” to “mild” pain during motion thereafter. We found no relationship between the presence of a nerve injury and the ability to perform a “touch-down.” Pain at rest had a small but statistically non-significant effect up to 13 weeks post-surgery. Pain during motion showed a statistically significant association up to 6 months post-surgery. Previous studies found no relationship between nerve laceration and return of finger mobility, but they did not distinguish between pain at rest and pain during motion.^16,34

The DASH has previously been used in studies of patients after flexor tendon injury.^{12,26,32,35–38} In a study by Wong et al.³⁹ patients at discharge after acute traumatic hand injuries achieved on average 14 (SD 15) DASH points. These results are comparable to our scores, as patients are typically discharged from therapy three months after flexor tendon surgery. The presence of a concomitant nerve injury was only significantly associated with the DASH scores at 13 weeks post-surgery, resulting in an average increase of 4.6 points.

We found a statistically significant relationship between pain during motion and the DASH up to one year after flexor tendon repair, while pain at rest did not show any association with hand function. Previous studies found generally low DASH scores after flexor tendon injuries,^{12,26,32,35–37} leaving little room for clinically important changes. For example, by scoring on average 17.0 points at week 13, achieving a clinically important change of 18.8 points⁴⁰ was not possible for our patients. Therefore, it is questionable if this is the right instrument to evaluate hand function after a flexor tendon injury.

Patient satisfaction with their rehabilitation outcomes is considered an essential metric of quality regarding the outcome after hand surgery.⁴¹ There is moderate evidence that pain and limitations in functioning influence patient satisfaction.⁴² In our cohort, pain during motion was statistically significantly associated with patient satisfaction up to one year post-surgery. We only found a significant relationship between pain at rest and satisfaction at week 26, but not at week 13. This finding shows that pain at rest at week 13 was tolerated but started to negatively impact the satisfaction of these few patients (3%, n = 5) who continued to suffer from pain at rest 6 months post-surgery.

Limitations

This study has several limitations. First, during the assessment of pain intensity, we evaluated pain in a unidimensional way. We could have measured pain with multidimensional scales, including pain duration and relief. Well-validated instruments, such as the McGill Pain Questionnaire,⁴³ are recommended for a more comprehensive pain assessment. However, we considered the VNRS as an appropriate instrument for flexor tendon patients, as it reflects sensory and not affective pain dimensions, the former being of most concern in acute post-surgery-related pain.⁴⁴ Second, we did not specifically assess whether patients were at risk of fear of movement (kinesiophobia), as there is evidence that pain could lead to kinesiophobia and ultimately to fear of tendon rupture.⁴⁵ Third, our measurement time points were fixed beforehand, therefore we don’t know if we captured pain when it was most elevated. Fourth, although the same pain medications were prescribed and administered by hand surgeons in all participating clinics, the intake of pain medication may have been influenced by the hand therapist’s and/or surgeon’s way of encouraging patients to take them. This may have produced a bias within our results. A further limitation was the exclusion of 44 patients due to too much missing data regarding pain values; this reduced our sample size. Finally, due to the retrospective nature of our study design, we did not calculate a-priori sample size. Performing a posthoc power analysis is more common for this type of study. However, it is debatable if post-hoc power analyses really can inform power for detecting significant results because the random component in the study disappears once data are collected, challenging the rationale for making such power estimates.⁴⁶

Conclusions

This study shows a significant relationship between pain during motion and pain medication intake, TAM, DASH scores and patients’ satisfaction. No relationship was identified between pain at rest and any of the outcome measurements, indicating that monitoring pain at rest may be unnecessary with this patient group. These findings notably indicate that pain during motion plays a crucial role in flexor tendon rehabilitation. Especially, as it can hinder the patient from actively moving the injured finger from week 1 onwards, thus, negatively affecting clinical outcomes. Our key recommendation is therefore to monitor pain during motion as opposed to pain at rest during the rehabilitation of flexor tendon injuries in zones 1-2.

The presence of a concomitant nerve injury increased post-surgical pain medication intake and was associated with worse TAM scores in week 6 and higher self-perceived disability in week 13. This relationship is in contrast with previous studies that found no association between nerve laceration and return of finger mobility, but they did not distinguish between pain at rest and pain during motion.^26,31,32 Therefore, future studies should state whether they assess pain at rest or during motion, the latter being required for future flexor tendon research.

Footnotes

Acknowledgements

We would like to thank the Hand Therapy and Hand Surgery Departments of the University Hospitals of Zurich, Bern and Basel for their support in data collection and the patients for their willingness to participate in this study. Many thanks go to Lieutenant Colonel Jamie Jamison, Royal Marines Commando, Royal Navy, for proof reading the final manuscript.

Declaration of conflicting interests

The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

Funding was provided by the SUVA, the Swiss national accident insurance. The funds were used to pay set-up and maintenance of the data cloud (REDCap) used in this multi-center study and the support of a statistician.

Ethical approval

This study was approved by the ethical committee of Zurich, Switzerland (BASEC-Nr. 2017–02267).

Informed consent

Written informed consent was given by all patients prior to study entry.

Guarantor

BT.

Contributorship

All authors designed the study and developed the protocol. VBF, BT and LS entered patient data into REDCap, VBF and BT extracted the data used for this study and analysed the data. VBF wrote the manuscript. All authors reviewed and edited the manuscript and approved the final version.

ORCID iDs

Vera Beckmann-Fries

Maurizio Calcagni

Alexandre Kaempfen

Esther Vögelin

Bernadette Tobler-Ammann

References

Manninen

Karjalainen

Määttä

, et al. Epidemiology of flexor tendon injuries of the hand in a Northern finnish population. Scand J Surg 2017; 106: 278–282.

Peters

Jha

Ross

. Rehabilitation following surgery for flexor tendon injuries of the hand. Cochrane Database Syst Rev 2021; 1: CD012479. Epub ahead of print 7 January 2017. DOI: 10.1002/14651858.CD012479

MacDermid

Valdes

Szekeres

, et al. The assessment of psychological factors on upper extremity disability: a scoping review. J Hand Ther 2018; 31: 511–523.

Newington

Lane

Holmes

, et al. Variation in patient information and rehabilitation regimens after flexor tendon repair in the United Kingdom. Hand Ther 2022: 175899832210896: 49–57.

Feitz

van Kooij

ter Stege

MHP

, et al. Closing the loop: a 10-year experience with routine outcome measurements to improve treatment in hand surgery. EFORT Open Rev 2021; 6: 439–450.

Pettengill

. The evolution of early mobilization of the repaired flexor tendon. J Hand Ther 2005; 18: 157–168.

Hand Surgery Quality Consortium, Ring

Akelman

. How should we use patient-reported outcome measures at the point of care in hand surgery? J Hand Surg Am 2021; 46: 1049–1056.

Raja

Carr

Cohen

, et al. The revised international association for the study of pain definition of pain: concepts, challenges, and compromises. Pain 2020; 161: 1976–1982.

Pogatzki-Zahn

Segelcke

Schug

. Postoperative pain—from mechanisms to treatment. PAIN Rep 2017; 2: e588.

10.

Glare

Aubrey

Myles

. Transition from acute to chronic pain after surgery. Lancet 2019; 393: 1537–1546.

11.

Rosberg

Carlsson

Höjgård

, et al. What determines the costs of repair and rehabilitation of flexor tendon injuries in zone II? a multiple regression analysis of data from Southern Sweden. J Hand Surg Br 2003; 28: 106–112.

12.

Trumble

Vedder

Seiler

, et al. Zone-II flexor tendon repair: a randomized prospective trial of active place-and-hold therapy compared with passive motion therapy. J Bone Joint Surg Am 2010; 92: 1381–1389.

13.

Edsfeldt

Eklund

Wiig

. Prognostic factors for digital range of motion after intrasynovial flexor tendon injury and repair: long-term follow-up on 273 patients treated with active extension-passive flexion with rubber bands. J Hand Ther 2019; 32: 328–333.

14.

Daluiski

, et al. The epidemiology of reoperation after flexor tendon repair. J Hand Surg Am 2012; 37: 919–924.

15.

Rigo

Røkkum

. Predictors of outcome after primary flexor tendon repair in zone 1, 2 and 3. J Hand Surg Eur Vol 2016; 41: 793–801.

16.

Keane

Stonner

Pet

. Does digital nerve injury affect range of motion recovery after zone 2 flexor tendon repair? HAND 2021: 155894472110031: 230–235.

17.

Tang

. Outcomes and evaluation of flexor tendon repair. Hand Clin 2013; 29: 251–259.

18.

Holdgate

Asha

Craig

, et al. Comparison of a verbal numeric rating scale with the visual analogue scale for the measurement of acute pain. Emerg Med 2003; 15: 441–446.

19.

Randall

Zhang

, et al. Establishing the minimal clinically important difference and substantial clinical benefit for the pain visual analog scale in a postoperative hand surgery population. J Hand Surg Am 2022; 47: 645–653.

20.

Sloman

Wruble

Rosen

, et al. Determination of clinically meaningful levels of pain reduction in patients experiencing acute postoperative pain. Pain Manag Nurs 2006; 7: 153–158.

21.

Karcioglu

Topacoglu

Dikme

, et al.

A systematic review of the pain scales in adults: which to use?

Am J Emerg Med 2018; 36: 707–714.

22.

Hudak

Amadio

Bombardier

. Development of an upper extremity outcome measure: the DASH (disabilities of the arm, shoulder, and head). Am. J. Ind. Med 1996; 29(6): 7.

23.

Cohen

. Statistical power analysis. Curr Dir Psychol Sci 1992; 1(3): 98–101.

24.

von Elm

Altman

Egger

, et al. The Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) statement: guidelines for reporting observational studies. Int J Surg 2014; 12: 1495–1499.

25.

Hah

Nwaneshiudu

Cramer

, et al. Acute pain predictors of remote postoperative pain resolution after hand surgery. Pain Ther 2021; 10: 1105–1119.

26.

Starnes

Saunders

Means

. Clinical outcomes of zone II flexor tendon repair depending on mechanism of injury. J Hand Surg Am 2012; 37: 2532–2540.

27.

Frueh

Kunz

Gravestock

, et al. Primary flexor tendon repair in zones 1 and 2: early passive mobilization versus controlled active motion. J Hand Surg Am 2014; 39: 1344–1350.

28.

Moriya

Yoshizu

Tsubokawa

, et al. Outcomes of flexor tendon repairs in zone 2 subzones with early active mobilization. J Hand Surg Eur Vol 2017; 42: 896–902.

29.

Zhou

Qing

, et al. Outcomes of the six-strand M-Tang repair for zone 2 primary flexor tendon repair in 54 fingers. J Hand Surg Eur Vol 2017; 42: 462–468.

30.

Giesen

Reissner

Besmens

, et al. Flexor tendon repair in the hand with the M-Tang technique (without peripheral sutures), pulley division, and early active motion. J Hand Surg Eur Vol 2018; 43: 474–479.

31.

Sade

İnanir

Şen

, et al. Rehabilitation outcomes in patients with early and two-stage reconstruction of flexor tendon injuries. J Phys Ther Sci 2016; 28: 2214–2219.

32.

Solomons

Barrow

, et al. Device for zone-II flexor tendon repair: a multicenter, randomized, blinded, clinical trial. J Bone Joint Surg Am 2005; 87: 923–935.

33.

Kelley

Shauver

Chung

. Management of acute postoperative pain in hand surgery: a systematic review. J Hand Surg Am 2015; 40: 1610–1619.e1.

34.

Edsfeldt

Rempel

Kursa

, et al. In vivo flexor tendon forces generated during different rehabilitation exercises. J Hand Surg Eur Vol 2015; 40: 705–710.

35.

Rajappa

Menon

Kumar

, et al. Early active motion protocol following triple kessler repair for flexor tendon injury. J Orthop Surg 2014; 22: 96–99.

36.

Schaller

Baer

. Motion-stable flexor tendon repair with the Mantero technique in the distal part of the fingers. J Hand Surg Eur Vol 2010; 35: 51–55.

37.

Natal-Albelo

Olivella

Paraliticci-Márquez

, et al. Functional and disability assessment among hispanics with zone 2 flexor tendon injuries: comparative study between flexor digitorum superficialis repair and flexor digitorum superficialis excision. J Am Acad Orthop Surg Glob Res Rev 2020; 4: e20.0008100081.

38.

Libberecht

Lafaire

Van Hee

. Evaluation and functional assessment of flexor tendon repair in the hand. Acta Chir Belg 2006; 106: 560–565.

39.

Wong

JYP

Fung

BKK

Chu

MML

, et al. The use of disabilities of the arm, shoulder, and hand questionnaire in rehabilitation after acute traumatic hand injuries. J Hand Ther 2007; 20: 49–55.

40.

Maia

MVdP

de Moraes

dos Santos

JBG

, et al. Minimal important difference after hand surgery: a prospective assessment for DASH, MHQ, and SF-12. SICOT-J 2016; 2: 32.

41.

Waljee

Curtin

. Quality assessment in hand surgery. Hand Clin 2014; 30: 329–334.

42.

Marks

Herren

Vliet Vlieland

TPM

, et al. Determinants of patient satisfaction after orthopedic interventions to the hand: a review of the literature. J Hand Ther 2011; 24: 303–312.e10.

43.

Melzack

. The McGill pain questionnaire: major properties and scoring methods. Pain 1975; 1: 277–299.

44.

Huber

Suman

Rendo

, et al. Dimensions of “unidimensional” ratings of pain and emotions in patients with chronic musculoskeletal pain. Pain 2007; 130: 216–224.

45.

Tuna

Oskay

. Fear of movement and its effects on hand function after tendon repair. Hand Surg Rehabil 2018; 37: 247–251.

46.

Zhang

Hedo

Rivera

, et al.

Post hoc power analysis: is it an informative and meaningful analysis?

Gen Psychiatr 2019; 32: e100069.