HAND/PERIPHERAL NERVE Outcomes Article
Measuring Outcomes and Determining LongTerm Disability after Revision Amputation for Treatment of Traumatic Finger and Thumb Amputation Injuries Aviram M. Giladi, M.D. Evan P. McGlinn, B.S. Melissa J. Shauver, M.P.H. Taylor P. Voice, B.S. Kevin C. Chung, M.D., M.S. Ann Arbor, Mich.
Background: Disability ratings after finger amputations are based on anatomical injury according to the American Medical Association’s Guides to the Evaluation of Permanent Impairment. These ratings determine disability and compensation, without considering validated outcomes measures. The authors hypothesize that patient-reported outcomes reflect function and health-related quality of life after traumatic finger amputations, and that Guides scoring does not accurately rate postamputation disability. Methods: Patients were classified by amputation: single finger, thumb, multifinger, or multifinger plus thumb. Eighty-four patients completed functional tests, the Jebsen-Taylor Hand Function Test, and patient-reported outcomes [Brief Michigan Hand Questionnaire (MHQ), Quick Disabilities of the Arm, Shoulder and Hand (DASH) questionnaire, and the Short Form-36 health-related quality-of-life questionnaire). Patients were given disability scores according to the Guides. Pearson correlations between outcomes metrics were calculated, and linear regression evaluated associations between amputation group, Guides score, and outcomes measures. Results: The Brief MHQ and Quick DASH questionnaires had significant correlation with functional tests, the Jebsen-Taylor test, and the physical component summary of Short Form-36. Only the Brief MHQ correlated with the mental component summary of the Short Form-36 (r = 0.29, p = 0.02). The Guides score only correlated with the Jebsen-Taylor test (r = 0.47, p < 0.001). Regression results indicate that the Brief MHQ, Quick DASH, and Guides score predict Jebsen-Taylor test score; however, amputation group and Guides score do not predict patient-reported outcomes. Conclusions: The American Medical Association Guides score represents anatomical and functional outcomes without addressing mental health and other components of disability. As a result, Guides scoring is inadequate for determining postamputation disability. In evaluating composite amputation outcomes, Brief Michigan Hand Questionnaire outperformed other metrics. (Plast. Reconstr. Surg. 134: 746e, 2014.) CLINICAL QUESTION/LEVEL OF EVIDENCE: Risk, II.
I
n the United States, approximately 45,000 people suffer traumatic finger amputations annually, a third of which are work-related.1–4 Treatment options include replantation of the amputated segment or revision of the amputation From the Department of Surgery, Section of Plastic Surgery, University of Michigan Health System, and the University of Michigan Medical School. Received for publication April 7, 2014; accepted May 13, 2014. Copyright © 2014 by the American Society of Plastic Surgeons DOI: 10.1097/PRS.0000000000000591
746e
stump (“revision amputation”). In the United States, 73 percent of thumb and 88 percent of finger traumatic amputations are treated with revision amputation.5 These injuries are also a Disclosure: The authors have no financial interest to declare in relation to the content of this article.
his work was supported by T THE PLASTIC SURGERY FOUNDATION.
www.PRSJournal.com
Volume 134, Number 5 • Finger and Thumb Amputation Outcomes concern worldwide where inadequate access to care in many regions leads to replantation rates below 5 percent.6 The burden of traumatic amputations, including acute management of injuries, rehabilitation, and long-term disability, exact a large toll on patients and contribute substantially to health care costs.7–10 Understanding the outcomes of revision amputations has important implications for treatment and resource allocation; however, these outcomes have been inadequately investigated and are poorly understood. The paradigm in hand surgery outcomes assessment has shifted to using patient-reported outcomes, including general health-related quality-of-life questionnaires and questionnaires that focus specifically on the upper extremity.11–13 Patient-reported outcomes accurately reflect recovery11,13–18; however, the few studies on revision amputation outcomes have focused only on strength and sensory measures.19–22 Although these measurements provide valuable information, patient-reported outcomes provide unique insights into a patient’s functional and emotional health state, and are more reliable and responsive to changes in the patient’s condition.12,13,15,18,23 Despite these known advantages of patientreported outcomes, disability ratings for finger amputation patients are made using non–patientreported outcomes metrics.24,25 These ratings are based on anatomical level of injury as outlined in the American Medical Association’s Guides to the Evaluation of Permanent Impairment, the standard in impairment and disability assessment.24 Although initially designed to rate impairment, the Guides are used by many states to determine long-term disability.25 This disability rating, although potentially inaccurate, then determines short- and longterm financial compensation and occupational responsibilities.22,25,26 As hand trauma outcomes assessments have evolved, patients with revision amputations have largely been left behind. Understanding of these patients’ outcomes has not advanced since Swanson’s seminal work over 50 years ago,19 yet these injuries remain common and the number of patients with revision amputations continues to grow. It is necessary to generate data from patientreported outcomes that more accurately reflect recovery to improve assessment and disability rating for traumatic finger amputation patients. In this study, we use functional measures, composite functional tests, and general and system-specific patient-reported outcomes to evaluate patients with revision finger and thumb amputations after traumatic injuries. These outcomes are compared
to ratings determined by the American Medical Association Guides. We hypothesize that systemspecific patient-reported outcomes will outperform functional metrics by accurately reporting functional and quality-of-life outcomes after revision finger and thumb amputations, and will more accurately reflect the disability outcome than the Guides rating.
PATIENTS AND METHODS Study Sample This is a single-center retrospective cohort study that received approval from the University of Michigan Institutional Review Board. Patients aged 18 years or older at the time of the study who were treated with revision amputations after thumb or finger amputation injury were recruited for participation. The initial patient list was obtained by a search through the University of Michigan Clinical Charge Capture program, which is searchable back to 1998. We began by searching for International Classification of Diseases, Ninth Revision, Clinical Modification finger and thumb traumatic amputation diagnosis codes (886.0, 886.1, 885.0, and 885.1). We then filtered with Current Procedural Terminology codes for revision amputation of a finger or thumb (26951 and 26952) to obtain our initial patient list. With these International Classification of Diseases, Ninth Revision, Clinical Modification and Current Procedural Terminology codes, only amputations distal to the metacarpophalangeal joint were included. We confirmed search codes and the patient list with the University of Michigan billing and coding specialists. We then filtered the list to exclude patients with their most recent amputation procedure less than 1 year before the study, so that recruited patients were likely to have completed all recovery and therapy by the time of evaluation. Each patient’s chart was then reviewed. We excluded patients who had any replanted digits, although we did include patients who underwent failed replantation that resulted in revision amputation. Patients with additional trauma to the same limb, concomitant diagnoses affecting the upper extremity (e.g., carpal tunnel, rheumatoid arthritis), injuries to the upper extremity that required extensive soft-tissue reconstruction, and any additional procedures on the upper extremities within the past year (even if unrelated to the amputation) were excluded.
747e
Plastic and Reconstructive Surgery • November 2014 Patient Recruitment From this list, trained research assistants began with patient recruitment (Fig. 1). When speaking to the patients, a script was used to ensure consistency. During this call, recruiters confirmed that patients interested in participating would be able to complete the questionnaires in English, had no recent procedures or injuries to the hands or upper extremities, and had no ongoing litigation regarding their injury or potential disability compensation. Patients were offered $50 for completing all outcomes assessments for the study, and were reminded that this visit was only for assessment purposes and they would not be seeing a provider. If during the conversation the patient expressed interest in making an appointment with a provider, they were then transferred to our clinic schedulers and not involved in our study. First recruitment attempts were made by means of phone calls. After each patient was called twice, we transitioned to e-mail (if an e-mail address was available) or postcards. As we continued to filter through the list, we made follow-up phone calls, and finally ended by sending a second round of postcards to those patients still on the list. In addition to these efforts, we listed this study on the public University of Michigan Clinical Studies recruitment Web site.
Fig. 1. Flow diagram for patient recruitment.
748e
Assessments Patients agreeing to participate in this study were scheduled to come to our research offices for assessment. During this visit, the patients did not interact with surgical providers, and trained research assistants performed assessments. Three validated patient-reported outcomes were used: the Brief Michigan Hand Questionnaire; the Quick Disabilities of the Arm, Shoulder and Hand (DASH) questionnaire; and the 36-Item ShortForm Health Survey. The Brief Michigan Hand Questionnaire is a responsive 12-item instrument with strong correlation to the Michigan Hand Questionnaire.27 The Brief Michigan Hand Questionnaire contains two items in each of six domains: (1) overall hand function, (2) activities of daily living, (3) pain, (4) work performance, (5) aesthetics, and (6) patient satisfaction. Scores range from 0 to 100, with higher scores indicating better performance. The Quick Disabilities of the Arm, Shoulder and Hand questionnaire is an 11-item version of the Disabilities of the Arm, Shoulder and Hand questionnaire and has been validated for use after hand trauma.28–30 The Quick DASH centers on evaluating function and activities of daily living. Scores range from 0 to 100, with 100 indicating the most disability. The 36-Item Short-Form Health Survey
Volume 134, Number 5 • Finger and Thumb Amputation Outcomes is a general health-related quality-of-life instrument.31 It has shown responsiveness in upper extremity disorders,16,32,33 and results give physical component summary and mental component summary scores, both scaled 0 to 100, with 100 being “best health.” Grip strength, two-point pinch, and threepoint pinch testing were also performed. A standard, adjustable-handle Jamar dynamometer was used to measure grip strength and a B&L pinch gauge (B&L Engineering, Santa Ana, Calif.) was used to measure pinch strength as recommended by the American Society of Hand Therapists.34 To test composite upper extremity function, we used the Jebsen-Taylor Hand Function Test, which has previously been used in the evaluation of finger amputation patients.35,36 The Jebsen-Taylor test for our study included six timed exercises that replicate key activities of daily living: card turning, picking up small objects, simulated feeding, stacking checkers, picking up large light-weight objects, and picking up large heavy objects. We excluded the writing component of the JebsenTaylor because of its strong dependence on handedness and because the results can be difficult to interpret.35,37 All patients were also given an impairment score based on the rating scheme outlined in the American Medical Association Guides.24 This system is based entirely on anatomical level of injury, with each component of lost anatomy translating to a percentage deficiency in function. For multiple finger injuries, the percentage loss from each finger is weighted and combined to give the total hand function loss (Fig. 2) (see example calculation in the Appendix24). Statistical Analysis Patients were stratified into four groups based on anatomical loss: (1) single finger amputation, (2) thumb amputation, (3) multiple finger amputations, and (4) multiple finger amputations including thumb. We also subclassified by whether the patient had a proximal amputation, which included any injury proximal to the distal interphalangeal joint in the fingers or interphalangeal joint in the thumb. Descriptive statistics were used to describe strength and functional metrics and patient-reported outcome results for each group. One-way analysis of variance with Scheffé correction for multiple comparisons was used to evaluate the differences in outcome measures across the groups. Additional subgroup analysis was performed between patients with proximal amputation and those without. We then calculated
Fig. 2. Method used in AMA Guides to determine percentage disability from finger amputations, based on level of amputation. DIP, distal interphalangeal joint; PIP, proximal interphalangeal joint; MCP, metacarpophalangeal joint. Adapted with permission from Swanson AB, Hagert CG, Swanson GD. Evaluation of impairment of hand function. J Hand Surg Am. 1983;8:709–722.
Pearson correlations between the patient-reported outcome scores, functional tests, Jebsen-Taylor test results, and American Medical Association Guides scores. We elected to use Pearson analysis rather than Spearman because Pearson analysis is more susceptible to the effect of outliers and therefore more sensitive to variation in results. Linear regression models were used to evaluate whether amputation group, Guides score, demographic factors, or other elements predicted Jebsen-Taylor results or patient-reported outcomes. In these models, we controlled for age, workers’ compensation status, sex, household income, and whether the injured hand was dominant or nondominant. All statistical work was done using Stata 12.0 (StataCorp., College Station, Texas).
RESULTS From the medical and billing records, we identified 633 patients with revision amputations between the years 1998 and 2012. Through our multiple recruitment methods, we enrolled 84 patients for this study (Fig. 1). Demographic information for the study participants is listed in
749e
750e
23 (27.4) 61 (72.6) 43 (51.2) 11 (13.1) 24 (28.6) 6 (7.1)
6.0 ± 2.1 7.0 ± 2.1 6.0 ± 2.0 5.6 ± 3.1 34.7 ± 13.3 39.4 ± 9.1 28.6 ± 14.1 32.1 ± 18.7 34.4 ± 5.6 41.2 ± 5.5 41.5 ± 10.6‡ 47.8 ± 14.5‡ 53.5 ± 9.2 57.2 ± 5.4 53.7 ± 10.2 61.1 ± 3.7 51.5 ± 10.0 52.2 ± 8.3 50.8 ± 8.5 48.0 ± 10.2 11.4 ± 14.6 9.1 ± 11.9 17.9 ± 18.0 18.6 ± 20.6
bMHQ, Brief Michigan Hand Questionnaire; QuickDASH, Quick Disabilities of the Arm, Shoulder and Hand questionnaire; SF-36, 36-Item Short-Form Health Survey; PCS, physical component summary of the SF-36; SF-36 MCS, mental component summary of the SF-36; JTT, Jebsen-Taylor Hand Function Test. *The raw scores that compose these mean scores were used in analysis of variance with Scheffé correction to evaluate whether differences between groups were significant. †Amputation groups: 1 = single finger only, 2 = thumb only, 3 = multiple fingers, 4 = multiple fingers and thumb. ‡Analysis of variance with Scheffé correction, significant result compared with amputation group 1 (p ≤ 0.05).
44 (53.7) 38 (46.3)
84.2 ± 15.1 81.7 ± 11.1 74.9 ± 19.0 73.2 ± 17.6
2 (2.4) 16 (19.0) 13 (15.5) 30 (35.7) 11 (13.1) 12 (14.3)
43 11 24 6
75 (89) 9 (11)
1 2 3 4
69 (82) 15 (18)
No.
93.3 ± 53.7 13–264
Mean ± SD QuickDASH Score
52.8 ± 16.9 21–83
Group†
GED, General Educational Development.
Value (%)
Mean ± SD bMHQ Score
Age, yr Mean ± SD Range Months since injury Mean Range Sex Male Female Race White Nonwhite Education level Less than high school graduate High school graduate or GED Vocational/technical school Some college or associate degree College graduate Professional or graduate school Annual household income
Measuring Outcomes and Determining LongTerm Disability after Revision Amputation for Treatment of Traumatic Finger and Thumb Amputation Injuries Aviram M. Giladi, M.D. Evan P. McGlinn, B.S. Melissa J. Shauver, M.P.H. Taylor P. Voice, B.S. Kevin C. Chung, M.D., M.S. Ann Arbor, Mich.
Background: Disability ratings after finger amputations are based on anatomical injury according to the American Medical Association’s Guides to the Evaluation of Permanent Impairment. These ratings determine disability and compensation, without considering validated outcomes measures. The authors hypothesize that patient-reported outcomes reflect function and health-related quality of life after traumatic finger amputations, and that Guides scoring does not accurately rate postamputation disability. Methods: Patients were classified by amputation: single finger, thumb, multifinger, or multifinger plus thumb. Eighty-four patients completed functional tests, the Jebsen-Taylor Hand Function Test, and patient-reported outcomes [Brief Michigan Hand Questionnaire (MHQ), Quick Disabilities of the Arm, Shoulder and Hand (DASH) questionnaire, and the Short Form-36 health-related quality-of-life questionnaire). Patients were given disability scores according to the Guides. Pearson correlations between outcomes metrics were calculated, and linear regression evaluated associations between amputation group, Guides score, and outcomes measures. Results: The Brief MHQ and Quick DASH questionnaires had significant correlation with functional tests, the Jebsen-Taylor test, and the physical component summary of Short Form-36. Only the Brief MHQ correlated with the mental component summary of the Short Form-36 (r = 0.29, p = 0.02). The Guides score only correlated with the Jebsen-Taylor test (r = 0.47, p < 0.001). Regression results indicate that the Brief MHQ, Quick DASH, and Guides score predict Jebsen-Taylor test score; however, amputation group and Guides score do not predict patient-reported outcomes. Conclusions: The American Medical Association Guides score represents anatomical and functional outcomes without addressing mental health and other components of disability. As a result, Guides scoring is inadequate for determining postamputation disability. In evaluating composite amputation outcomes, Brief Michigan Hand Questionnaire outperformed other metrics. (Plast. Reconstr. Surg. 134: 746e, 2014.) CLINICAL QUESTION/LEVEL OF EVIDENCE: Risk, II.
I
n the United States, approximately 45,000 people suffer traumatic finger amputations annually, a third of which are work-related.1–4 Treatment options include replantation of the amputated segment or revision of the amputation From the Department of Surgery, Section of Plastic Surgery, University of Michigan Health System, and the University of Michigan Medical School. Received for publication April 7, 2014; accepted May 13, 2014. Copyright © 2014 by the American Society of Plastic Surgeons DOI: 10.1097/PRS.0000000000000591
746e
stump (“revision amputation”). In the United States, 73 percent of thumb and 88 percent of finger traumatic amputations are treated with revision amputation.5 These injuries are also a Disclosure: The authors have no financial interest to declare in relation to the content of this article.
his work was supported by T THE PLASTIC SURGERY FOUNDATION.
www.PRSJournal.com
Volume 134, Number 5 • Finger and Thumb Amputation Outcomes concern worldwide where inadequate access to care in many regions leads to replantation rates below 5 percent.6 The burden of traumatic amputations, including acute management of injuries, rehabilitation, and long-term disability, exact a large toll on patients and contribute substantially to health care costs.7–10 Understanding the outcomes of revision amputations has important implications for treatment and resource allocation; however, these outcomes have been inadequately investigated and are poorly understood. The paradigm in hand surgery outcomes assessment has shifted to using patient-reported outcomes, including general health-related quality-of-life questionnaires and questionnaires that focus specifically on the upper extremity.11–13 Patient-reported outcomes accurately reflect recovery11,13–18; however, the few studies on revision amputation outcomes have focused only on strength and sensory measures.19–22 Although these measurements provide valuable information, patient-reported outcomes provide unique insights into a patient’s functional and emotional health state, and are more reliable and responsive to changes in the patient’s condition.12,13,15,18,23 Despite these known advantages of patientreported outcomes, disability ratings for finger amputation patients are made using non–patientreported outcomes metrics.24,25 These ratings are based on anatomical level of injury as outlined in the American Medical Association’s Guides to the Evaluation of Permanent Impairment, the standard in impairment and disability assessment.24 Although initially designed to rate impairment, the Guides are used by many states to determine long-term disability.25 This disability rating, although potentially inaccurate, then determines short- and longterm financial compensation and occupational responsibilities.22,25,26 As hand trauma outcomes assessments have evolved, patients with revision amputations have largely been left behind. Understanding of these patients’ outcomes has not advanced since Swanson’s seminal work over 50 years ago,19 yet these injuries remain common and the number of patients with revision amputations continues to grow. It is necessary to generate data from patientreported outcomes that more accurately reflect recovery to improve assessment and disability rating for traumatic finger amputation patients. In this study, we use functional measures, composite functional tests, and general and system-specific patient-reported outcomes to evaluate patients with revision finger and thumb amputations after traumatic injuries. These outcomes are compared
to ratings determined by the American Medical Association Guides. We hypothesize that systemspecific patient-reported outcomes will outperform functional metrics by accurately reporting functional and quality-of-life outcomes after revision finger and thumb amputations, and will more accurately reflect the disability outcome than the Guides rating.
PATIENTS AND METHODS Study Sample This is a single-center retrospective cohort study that received approval from the University of Michigan Institutional Review Board. Patients aged 18 years or older at the time of the study who were treated with revision amputations after thumb or finger amputation injury were recruited for participation. The initial patient list was obtained by a search through the University of Michigan Clinical Charge Capture program, which is searchable back to 1998. We began by searching for International Classification of Diseases, Ninth Revision, Clinical Modification finger and thumb traumatic amputation diagnosis codes (886.0, 886.1, 885.0, and 885.1). We then filtered with Current Procedural Terminology codes for revision amputation of a finger or thumb (26951 and 26952) to obtain our initial patient list. With these International Classification of Diseases, Ninth Revision, Clinical Modification and Current Procedural Terminology codes, only amputations distal to the metacarpophalangeal joint were included. We confirmed search codes and the patient list with the University of Michigan billing and coding specialists. We then filtered the list to exclude patients with their most recent amputation procedure less than 1 year before the study, so that recruited patients were likely to have completed all recovery and therapy by the time of evaluation. Each patient’s chart was then reviewed. We excluded patients who had any replanted digits, although we did include patients who underwent failed replantation that resulted in revision amputation. Patients with additional trauma to the same limb, concomitant diagnoses affecting the upper extremity (e.g., carpal tunnel, rheumatoid arthritis), injuries to the upper extremity that required extensive soft-tissue reconstruction, and any additional procedures on the upper extremities within the past year (even if unrelated to the amputation) were excluded.
747e
Plastic and Reconstructive Surgery • November 2014 Patient Recruitment From this list, trained research assistants began with patient recruitment (Fig. 1). When speaking to the patients, a script was used to ensure consistency. During this call, recruiters confirmed that patients interested in participating would be able to complete the questionnaires in English, had no recent procedures or injuries to the hands or upper extremities, and had no ongoing litigation regarding their injury or potential disability compensation. Patients were offered $50 for completing all outcomes assessments for the study, and were reminded that this visit was only for assessment purposes and they would not be seeing a provider. If during the conversation the patient expressed interest in making an appointment with a provider, they were then transferred to our clinic schedulers and not involved in our study. First recruitment attempts were made by means of phone calls. After each patient was called twice, we transitioned to e-mail (if an e-mail address was available) or postcards. As we continued to filter through the list, we made follow-up phone calls, and finally ended by sending a second round of postcards to those patients still on the list. In addition to these efforts, we listed this study on the public University of Michigan Clinical Studies recruitment Web site.
Fig. 1. Flow diagram for patient recruitment.
748e
Assessments Patients agreeing to participate in this study were scheduled to come to our research offices for assessment. During this visit, the patients did not interact with surgical providers, and trained research assistants performed assessments. Three validated patient-reported outcomes were used: the Brief Michigan Hand Questionnaire; the Quick Disabilities of the Arm, Shoulder and Hand (DASH) questionnaire; and the 36-Item ShortForm Health Survey. The Brief Michigan Hand Questionnaire is a responsive 12-item instrument with strong correlation to the Michigan Hand Questionnaire.27 The Brief Michigan Hand Questionnaire contains two items in each of six domains: (1) overall hand function, (2) activities of daily living, (3) pain, (4) work performance, (5) aesthetics, and (6) patient satisfaction. Scores range from 0 to 100, with higher scores indicating better performance. The Quick Disabilities of the Arm, Shoulder and Hand questionnaire is an 11-item version of the Disabilities of the Arm, Shoulder and Hand questionnaire and has been validated for use after hand trauma.28–30 The Quick DASH centers on evaluating function and activities of daily living. Scores range from 0 to 100, with 100 indicating the most disability. The 36-Item Short-Form Health Survey
Volume 134, Number 5 • Finger and Thumb Amputation Outcomes is a general health-related quality-of-life instrument.31 It has shown responsiveness in upper extremity disorders,16,32,33 and results give physical component summary and mental component summary scores, both scaled 0 to 100, with 100 being “best health.” Grip strength, two-point pinch, and threepoint pinch testing were also performed. A standard, adjustable-handle Jamar dynamometer was used to measure grip strength and a B&L pinch gauge (B&L Engineering, Santa Ana, Calif.) was used to measure pinch strength as recommended by the American Society of Hand Therapists.34 To test composite upper extremity function, we used the Jebsen-Taylor Hand Function Test, which has previously been used in the evaluation of finger amputation patients.35,36 The Jebsen-Taylor test for our study included six timed exercises that replicate key activities of daily living: card turning, picking up small objects, simulated feeding, stacking checkers, picking up large light-weight objects, and picking up large heavy objects. We excluded the writing component of the JebsenTaylor because of its strong dependence on handedness and because the results can be difficult to interpret.35,37 All patients were also given an impairment score based on the rating scheme outlined in the American Medical Association Guides.24 This system is based entirely on anatomical level of injury, with each component of lost anatomy translating to a percentage deficiency in function. For multiple finger injuries, the percentage loss from each finger is weighted and combined to give the total hand function loss (Fig. 2) (see example calculation in the Appendix24). Statistical Analysis Patients were stratified into four groups based on anatomical loss: (1) single finger amputation, (2) thumb amputation, (3) multiple finger amputations, and (4) multiple finger amputations including thumb. We also subclassified by whether the patient had a proximal amputation, which included any injury proximal to the distal interphalangeal joint in the fingers or interphalangeal joint in the thumb. Descriptive statistics were used to describe strength and functional metrics and patient-reported outcome results for each group. One-way analysis of variance with Scheffé correction for multiple comparisons was used to evaluate the differences in outcome measures across the groups. Additional subgroup analysis was performed between patients with proximal amputation and those without. We then calculated
Fig. 2. Method used in AMA Guides to determine percentage disability from finger amputations, based on level of amputation. DIP, distal interphalangeal joint; PIP, proximal interphalangeal joint; MCP, metacarpophalangeal joint. Adapted with permission from Swanson AB, Hagert CG, Swanson GD. Evaluation of impairment of hand function. J Hand Surg Am. 1983;8:709–722.
Pearson correlations between the patient-reported outcome scores, functional tests, Jebsen-Taylor test results, and American Medical Association Guides scores. We elected to use Pearson analysis rather than Spearman because Pearson analysis is more susceptible to the effect of outliers and therefore more sensitive to variation in results. Linear regression models were used to evaluate whether amputation group, Guides score, demographic factors, or other elements predicted Jebsen-Taylor results or patient-reported outcomes. In these models, we controlled for age, workers’ compensation status, sex, household income, and whether the injured hand was dominant or nondominant. All statistical work was done using Stata 12.0 (StataCorp., College Station, Texas).
RESULTS From the medical and billing records, we identified 633 patients with revision amputations between the years 1998 and 2012. Through our multiple recruitment methods, we enrolled 84 patients for this study (Fig. 1). Demographic information for the study participants is listed in
749e
750e
23 (27.4) 61 (72.6) 43 (51.2) 11 (13.1) 24 (28.6) 6 (7.1)
6.0 ± 2.1 7.0 ± 2.1 6.0 ± 2.0 5.6 ± 3.1 34.7 ± 13.3 39.4 ± 9.1 28.6 ± 14.1 32.1 ± 18.7 34.4 ± 5.6 41.2 ± 5.5 41.5 ± 10.6‡ 47.8 ± 14.5‡ 53.5 ± 9.2 57.2 ± 5.4 53.7 ± 10.2 61.1 ± 3.7 51.5 ± 10.0 52.2 ± 8.3 50.8 ± 8.5 48.0 ± 10.2 11.4 ± 14.6 9.1 ± 11.9 17.9 ± 18.0 18.6 ± 20.6
bMHQ, Brief Michigan Hand Questionnaire; QuickDASH, Quick Disabilities of the Arm, Shoulder and Hand questionnaire; SF-36, 36-Item Short-Form Health Survey; PCS, physical component summary of the SF-36; SF-36 MCS, mental component summary of the SF-36; JTT, Jebsen-Taylor Hand Function Test. *The raw scores that compose these mean scores were used in analysis of variance with Scheffé correction to evaluate whether differences between groups were significant. †Amputation groups: 1 = single finger only, 2 = thumb only, 3 = multiple fingers, 4 = multiple fingers and thumb. ‡Analysis of variance with Scheffé correction, significant result compared with amputation group 1 (p ≤ 0.05).
44 (53.7) 38 (46.3)
84.2 ± 15.1 81.7 ± 11.1 74.9 ± 19.0 73.2 ± 17.6
2 (2.4) 16 (19.0) 13 (15.5) 30 (35.7) 11 (13.1) 12 (14.3)
43 11 24 6
75 (89) 9 (11)
1 2 3 4
69 (82) 15 (18)
No.
93.3 ± 53.7 13–264
Mean ± SD QuickDASH Score
52.8 ± 16.9 21–83
Group†
GED, General Educational Development.
Value (%)
Mean ± SD bMHQ Score
Age, yr Mean ± SD Range Months since injury Mean Range Sex Male Female Race White Nonwhite Education level Less than high school graduate High school graduate or GED Vocational/technical school Some college or associate degree College graduate Professional or graduate school Annual household income
