Predicting Functional Recovery and Return to Work After Mutilating Hand Injuries: Usefulness of Campbell's Hand Injury Severity Score

Predicting Functional Recovery and Return to Work After Mutilating Hand Injuries: Usefulness of Campbell's Hand Injury Severity Score

SCIENTIFIC ARTICLE Predicting Functional Recovery and Return to Work After Mutilating Hand Injuries: Usefulness of Campbell’s Hand Injury Severity Sc...

638KB Sizes 0 Downloads 63 Views

SCIENTIFIC ARTICLE

Predicting Functional Recovery and Return to Work After Mutilating Hand Injuries: Usefulness of Campbell’s Hand Injury Severity Score Hironori Matsuzaki, MD, Hiroko Narisawa, MD, Hitoshi Miwa, MD, Satoshi Toishi, MD

Purpose We investigated the effect of severity and form of mutilating hand injuries on functional recovery and return to work, and the usefulness of an injury-severity score in predicting these outcomes. Methods We reexamined patients in whom finger amputation or near amputation had been treated with replantation or revascularization at our institution at least 3 years earlier. We evaluated radiographs, sensory recovery, finger range of motion, skin temperature, and current overall hand function to calculate Tamai and Quick Disabilities of the Arm, Shoulder, and Hand scores, and recorded return-to-work status, duration of treatment, and time away from work. We determined injury level, number of involved fingers, type of injury, and Campbell’s Hand Injury Severity score (HISS) from the medical record. Results The average age of the 50 enrolled patients was 43 years (range, 18 – 69 years); average follow-up was 7.8 years (range, 3.1–15.3 years). More proximal injuries, more involved fingers, and more complicated injury predicted poorer functional recovery. HISS was highly correlated with Tamai’s score (r ⫽ – 0.77; p⬍.001) and moderately correlated with the Quick Disabilities of the Arm, Shoulder, and Hand score (r ⫽ 0.39; p ⫽ .009). HISS was only moderately correlated with length of treatment (r ⫽ 0.32; p⬍.05) and with time away from work (r ⫽ 0.34; p⬍.05). Mean HISS among the 3 return-to-work groups differed remarkably. When HISS was ⬍50, 11 of 12 patients returned to their original jobs; when it was between 50 and 150, 17 of 23 patients were able to return; and when it was ⬎150, only 4 of 15 were able to return to work. Conclusions Given the current surgical care of patients with mutilated hand injuries, HISS determined at the time of injury can adequately predict functional outcome and return to work status. Larger studies will be necessary to validate these findings. (J Hand Surg 2009;34A:880– 885. © 2009 Published by Elsevier Inc. on behalf of the American Society for Surgery of the Hand.) Type of study/level of evidence Prognostic III. Key words Amputation, near amputation, functional outcome, return to work, hand. EVERAL SCORING SYSTEMS are regularly used to guide surgical decision making for trauma patients.1–3 These scoring systems were developed to inform decisions about whether the injured extremity

S

FromtheDepartmentofOrthopaedicSurgery,TsubameRosaiHospital;theNiigataHandSurgeryFoundation; and the Department of Orthopaedic Surgery, Niigata Prefectural Shibata Hospital, Niigata, Japan. Received for publication September 16, 2008; accepted in revised form February 6, 2009. Funding for this study was provided by Japan Labour Health and Welfare Organization.

880 䉬 ©  Published by Elsevier, Inc. on behalf of the ASSH.

should be salvaged, and are also useful for assessing prognosis. Although such scoring systems for the feet and legs have been well established, for arm and hand injuries, especially mutilating hand injuries, there is no Corresponding author: Hironori Matsuzaki, MD, Department of Orthopaedic Surgery, Tsubame Rosai Hospital, 633 Sawatari, Tsubame-shi, Niigata 959-1228 Japan; e-mail: [email protected]. 0363-5023/09/34A05-0011$36.00/0 doi:10.1016/j.jhsa.2009.02.009

881

PREDICTING RECOVERY FROM SEVERE HAND INJURY

established or widely accepted scoring system and no systematic, validated method for predicting eventual recovery. Moreover, the effects of specific injury characteristics of mutilating hand injuries on functional and work-related outcomes have not been well described. Recently, Campbell and Kay4 developed the Hand Injury Severity score (HISS) to quantify injury severity and to allow treatment results from different institutions to be compared with a common baseline. A hand injury score that could predict functional outcome and job return might be useful in referring patients to appropriate institutions for more advanced treatment, and also of interest to patients and their employers. We sought to determine the effect of the form and severity of mutilating hand injuries on functional outcome and return to work, and to verify the usefulness of the HISS in predicting these end points. Here, we report the results of a study in which we evaluated patients after recovery and then retrospectively scored the degree of hand injury with the HISS to determine its predictive value. MATERIALS AND METHODS In this retrospective study, we reviewed the records of patients seen at our institution between October 1990 and August 2003 for work-related, mutilating hand injuries involving amputation or near amputation that required replantation or revascularization, or injuries that required microsurgical free-tissue transfer reconstruction. Of 82 eligible patients with a postoperative follow-up of at least 3 years, 50 (40 men) agreed to participate in the study and were examined directly to measure functional recovery. The average age at the time of injury was 43 years (range, 18 – 69 years). All patients were manual workers in factories, construction, or carpentry, and all injuries were work-related accidents that were treated under workers’ accident compensation insurance. The number of injured fingers ranged from 1 to 7, and all 3 forms of injury were represented (Table 1). Also included in this series were 8 wraparound flaps and 2 vascularized toe joint transfers. The study was approved by the institutional review board of our institution, and written informed consent was obtained from all participants. We evaluated current overall hand function with Tamai’s score5 and with the 11-item Quick Disabilities of the Arm, Shoulder, and Hand (DASH) score.6 Tamai’s score assigns point values to range of motion (up to 20 possible points), activities of daily living (20 possible points), sensation (20 possible points), subjective symptoms (10 possible points), appearance (10

TABLE 1. Characteristics of 50 Patients With Mutilating Hand Injuries Characteristic

Value

Gender (n) Men

40

Women

10

Mean (range) age (y)

43 (18–69)

Mean (range) follow-up (y)

7.8 (3.1–15.3)

Side of injury (n) Right

23

Left

25

Bilateral

2

Number of injured fingers (n) 1

25

2

11

3

5

4

4

5

4

6

0

7

1

Number of amputations

64 fingers, 31 patients 32 replantation, 32 stump plasty

Number of near amputations

41 fingers, 19 patients 35 revascularization, 6 stump plasty

Level of injury (patients, n) Zone I

1

Zone II

7

Zone III

16

Zone IV

13

Zone V

13

Type of injury (patients, n) Clean

3

Crush

42

Avulsion Other

3 2 (degloving; heat press)

possible points), and patient satisfaction (20 possible points). Scores range from 0 (worst function) to 100 (best function) and are categorized as poor (0 –39), fair (40 –59), good (60 –70), and excellent (80 –100). The Quick DASH score is based on a patient’s responses to 11 questions for activities of daily living and job and hobbies (sports or playing musical instruments), each graded from 1 (no difficulty) to 5 (unable to perform the movement). Scores range from 0 (best function) to 300 (worst function).

JHS 䉬 Vol A, May–June 

882

PREDICTING RECOVERY FROM SEVERE HAND INJURY

FIGURE 1: A, B, C Tamai’s scores for current hand function of 50 patients with mutilating hand injuries, by injury characteristic. Boxes indicate medians and interquartile ranges of each group; vertical lines indicate 10th and 90th percentiles; circles indicate data falling beyond the 10th and 90th percentiles. D P10, 10th percentile; P25, 25th percentile; P50, 50th percentile (median); P75, 75th percentile. A *p⬍.05, Steel-Dwass test.

From the medical record, we recorded the number of injured fingers, level of injury, form of injury, and duration of treatment. The level of injury was classified as zone I to V (from distal to proximal) with Tamai’s classification,7 and the form of injury was classified as a clean (sharp) cut, crush, or avulsion. Each patient reported time away from work. With information from the medical record, we scored the overall severity of the injury with the HISS.4 This score is created by assigning values for injuries to the integument (skin), skeleton (bone), motor (tendon), and nerves of each finger. The values are determined by the area and extent of injury to each involved finger. The sum of these values for each finger is multiplied by a constant assigned to each finger. The sum for all fingers is the HISS, which ranges from 0 (no injury) to 826 (theoretically assumed maximum severity); Campbell and Kay4 originally categorized the injuries as minor (0 –20), moderate (21–50), severe (51–100), and major (⬎100). Statistical methods We analyzed differences in overall hand function score (Tamai and Quick DASH scores) within different injurylevel groups (zone I–V), different finger-number groups (from 1 to 7), and different injury-form groups (clean cut, crush, and avulsion) with the Steel-Dwass test, a

nonparametric test for comparing 3 or more groups. Correlations among HISS and Tamai’s scores, Quick DASH scores, duration of treatment, and time away from work were assessed with Spearman’s rank correlation coefficient. We also used the Steel-Dwass test to compare the mean HISS of patients in the 3 job-return levels (return to original job, change jobs, or unable to work). We set alpha to 0.05 to adjust for multiple comparisons (Steel-Dwass test) and to 0.05 for correlation analysis. All tests were 2-tailed. The statistical analysis software package Statview5.0 (SAS Institute. Inc., Cary, NC) and free statistical software “R” were used in the analysis. RESULTS Injury characteristics and functional recovery Tamai and Quick DASH scores worsened as the level of injury became more proximal or as more fingers were injured, indicating a higher level of impairment and disability. This relationship was clearer in Tamai scores than in Quick DASH scores (Figs. 1, 2). Here, impairment is defined as an impediment to basic hand and finger function—that is, finger range of motion and sensory function—and the disability is defined as an impediment to an applied or acquired hand-and-finger

JHS 䉬 Vol A, May–June 

PREDICTING RECOVERY FROM SEVERE HAND INJURY

883

FIGURE 2: A, B, C Quick DASH scores for current hand function of 50 patients with mutilating hand injuries, by injury characteristic. Boxes indicate medians and interquartile ranges of each group; vertical lines indicate 10th and 90th percentiles; circles indicate data falling beyond the 10th and 90th percentiles. D P10, 10th percentile; P25, 25th percentile; P50, 50th percentile (median); P75, 75th percentile. A, B, C p⬎.05 for all comparisons between groups (Steel-Dwass test).

FIGURE 3: Correlations between HISS and functional outcome as measured by Tamai and Quick DASH scores. A The correlation between the HISS and Tamai score was – 0.77, which indicates a strong inverse relationship (p⬍.001). B Although the correlation between the HISS and Quick DASH scores was statistically significant (p ⫽ .009), the correlation coefficient was only 0.39, which indicates a moderate correlation.

function, such as writing, buttoning, and playing a musical instrument. As the form of injury became more severe (that is, clean injuries were less severe than crush injuries, which were less severe than avulsion injuries), both Tamai and Quick DASH scores worsened. This relationship was also clearer in Tamai scores than in Quick DASH scores (Figs. 1, 2). HISS score and final function The correlation between HISS and Tamai’s score was –0.77, which indicates a strong negative correlation (p⬍.001) between a higher injury score and poorer final function. The correlation between HISS and Quick DASH

score was statistically significant (p⬍.01) but only 0.39, which indicates a moderate correlation (Fig. 3). Differences in functional scores by level of return to job Median Tamai’s score (in which 100 is the best function) was 78 for those returning to their original jobs, 67 for those changing jobs, and 39 for those unable to work. All scores were statistically different from each other. Median DASH score (in which 0 is the best function) was 15 for those returning to their original jobs, 15 for those changing jobs, and 49 for those unable to work. Only the difference between those unable to work and the other 2 groups was statistically significant (Steel-Dwass test).

JHS 䉬 Vol A, May–June 

884

PREDICTING RECOVERY FROM SEVERE HAND INJURY

FIGURE 5: Relationship between length of treatment and return to work. For patients whose treatment lasted ⬍14 months, 27 of 38 were able to return to their original jobs. However, for patients whose treatment lasted ⱖ14 months, only 4 of 12 were able to return to their original jobs, which indicates an association between longer treatment and lower chances of returning to the original job.

FIGURE 4: Distribution of HISS for 50 patients with mutilating hand injuries, by level of return to job. A As the job-return level worsened, from original job to job change, to unable to work, the HISS became higher and the difference between the group with the original job and that with a changed job became statistically significant. Boxes indicate medians and interquartile ranges of each group; vertical lines indicate 10th and 90th percentiles; circles indicate data falling beyond the 10th and 90th percentiles. *p⬍.05, Steel-Dwass test. B Analysis of job-return level demonstrates that when the HISS was ⬍50, 11 of 12 patients were able to return to their original job. When HISS was between 50 and 150, 17 of 23 (74%) could return to their original jobs. When HISS was ⬎150, only 4 of 14 patients returned to their original jobs. Original, original job; change, job change; unable, unable to work.

time away from work (15 vs 9 months). The correlation between HISS and treatment period was ␳ ⫽ 0.32 (p ⫽ .03), and that between HISS and time away from work was ␳ ⫽ 0.34 (p ⫽ .02), which indicates weak relationships. For patients whose treatment lasted ⬍14 months, 27 of the 38 could return to their original jobs. However, for patients whose treatment lasted ⱖ14 months, only 4 of the 12 could return to their original jobs, which indicates an association between longer treatment and lower chances of returning to the original job (Fig. 5).

HISS score and level of return to job As the level of return to job worsened from original job to job change, to unable to work, the HISS became higher and the difference between the original job group and change job group became statistically significant (Fig. 4A). Analysis of the level of return to job revealed that when the HISS was ⬍50, 11 of 12 patients were able to return to their original job. When HISS was between 50 and 150, 17 of 23 returned to their original jobs. When HISS was ⬎150, only 4 of 14 patients returned to their original jobs (Fig. 4B). The duration of treatment, defined as the period from the initial injury to the termination of clinical follow-up, ranged from 2 to 32 months (mean, 11 months); the length of time away from work ranged from 2 to 39 months (mean, 10 months). The duration of treatment for the 10 free-tissue-transfer patients was longer than that of the 40 other patients (18 vs 9 months), as was

DISCUSSION Our results showed that amputation or near amputation at a more proximal level, more injured fingers, and more complicated forms of injury (from clean to crush to avulsion) predicted poorer functional recovery. This relationship was more apparent in Tamai scores than in Quick DASH scores. Moreover, as an overall objective scoring system of injury severity that contains many injury variables, the HISS was more closely correlated with Tamai’s scores (Spearman’s ␳ ⫽ – 0.77) than with the Quick DASH score (Spearman’s ␳ ⫽ 0.38). Tamai’s score consists of both objective criteria, such as range of motion and sensory evaluation, and subjective criteria, such as working ability and patient satisfaction. Furthermore, both doctors and patients provide information about impairment of basic finger function and disabilities at work and in the activities in daily living.5 In contrast, the Quick DASH score consists of only subjective evaluations by patients about their disability, for applied (or acquired) work: that is, activities

JHS 䉬 Vol A, May–June 

PREDICTING RECOVERY FROM SEVERE HAND INJURY

of daily living and satisfaction with job and hobbies. All those evaluations are related to arm, shoulder, and hand function. Thus, we believe that the objective scoring of injury severity by the HISS explains its closer correlation with the Tamai score, which also contains many objective and quantitative outcomes, and its weaker correlation with the Quick DASH scores, which contains only subjective evaluations. Ueba8 and Oosterom et al.9 reported that evaluations of impairment and disability do not always coincide with evaluations of the functional recovery of severely injured hands. However, the importance of the Quick DASH score should not be underestimated because it is a functional evaluation from the patient’s point of view. Instead, Quick DASH scores should supplement Tamai scores. When evaluating functional recovery from a mutilating hand injury, we should use both the Tamai score, which mainly contains objective and quantitative criteria assessed by physicians, and the Quick DASH score, which contains only subjective criteria assessed by patients. The HISS accurately predicted return to work but was not highly correlated with the length of treatment or with time away from work. We speculate that the length of the treatment period and time away from work are affected by factors not included in the HISS, such as the specifics of a given occupation or the skill of the worker before the injury. These relationships may be worth investigating further. Although the HISS is useful for predicting recovery status, it has several shortcomings. Scores are calculated for injury to the integument, skeleton, tendons, and nerves. However, vascular injury, which existed in all of our patients, is not evaluated.10 Moreover, the sum for each finger is multiplied by a constant that is said to be proportional to the functional contribution of each finger (thumb ⫽ 6, index ⫽ 2, middle ⫽ 3, ring ⫽ 3, and little ⫽ 2), and then the overall score is determined by summing values for each injured finger. However, the appropriateness of these constants has not been fully justified and may need more discussion.11 Taking these scoring methods into account, for the HISS, we should reconsider scoring vascular injury and the appropriateness of the constant assigned to each finger, to establish an unbiased and reasonable scoring system for a variety of hand injuries. In this study, we evaluated patients with the most severe form of hand injuries (amputation and near amputation), whereas the full range of hand trauma contains a wider variation of severity. We believe that the HISS could be refined by investigating its relationship

885

with evaluations of impairment (Tamai score), disability (Tamai score and the Quick DASH score), and job-return level for patients with a wider range of injury and a variety of personal and employment backgrounds from multiple institutions. Given the current surgical care of patients with mutilated hand injuries, HISS determined at the time of injury can adequately predict functional outcome and return to work status. Larger studies will be necessary to validate these findings. Limitation of the study Although functional recovery was assessed directly for each patient, injury severity was assessed retrospectively from the medical record. We have no reason to believe that the records were inaccurate, but it would have been preferable to assign an HISS at the initial examination as part of a prospective study. We did not have information on the psychological implications of these injuries, so we could not assess the relationship between psychological issues and job return and residual symptoms. The entirely subjective DASH score reflects patients’ psychological issues more than Tamai’s score, 70% of which is based on objective measurements. REFERENCES 1. Howe HR, Poole GV, Hansen KJ, Clerk T, Plonk GW, Koman LA, et al. Salvage of lower extremities following combined orthopaedic and vascular trauma: a predictive index. Am Surg 1987;53:205–208. 2. Johansen K, Daines M, Howey T, Helfet D, Hansen ST. Objective criteria accurately predict amputation following lower extremity trauma. J Trauma 1990;30:568 –572. 3. Russell WL, Sailors DM, Whittle TB, Fisher DF, Burns RP. Limb salvage versus traumatic amputation: a decision based on seven-part predictive index. Ann Surg 1991;213:473– 480. 4. Campbell DA, Kay SPJ. The hand injury severity scoring system. J Hand Surg 1996;21B:295–298. 5. Tamai S. Twenty years’ experience of limb replantation: review of 293 upper extremity replants. J Hand Surg 1982;7A:549 –556. 6. Imaeda T, Toh S, Wada T, Uchiyama S, Okinaga S, Kusunose K. Validation of the Japanese society for surgery of the hand version of the quick disability of the arm, shoulder, and hand (quick DASHJSSH) questionnaire. J Orthop Sci 2006;11:248 –253. 7. Tami S, Hori Y, Tatsumi Y, Okuda H, Nakamura Y, Sakamoto H, et al. Microvascular anastomosis and its application on the replantation of amputated digits and hands. Clin Orthop Relat Res 1978;133: 106 –121. 8. Ueba Y. Functional evaluation of the hand from the viewpoint of impairment, disability and handicap. J Jpn Surg Hand 1991;5:893– 895. 9. Oosterom FJT, Ettema AM, Mulder PGH, Hovius SER. Impairment and disability after severe hand injuries with multiple phalangeal fractures. J Hand Surg 2007;32A:91–95. 10. Saxena P, Cutler L, Feldberg L. Assessment of the severity of hand injuries using “hand injury severity score,” and its correlation with the functional outcome. Injury 2003;35:511–516. 11. Giacomozzi C, Giansanti D, Morelli S, Maccioni G, Macellari V. Objective analysis of finger function. Hand Clin 2003;19:421– 430.

JHS 䉬 Vol A, May–June 