- Email: [email protected]
Short-term mortality and risk factors associated with reoperation after emergency amputation of the lower limbs
Journal of Orthopaedic Science xxx (xxxx) xxx
Contents lists available at ScienceDirect
Journal of Orthopaedic Science journal homepage: http://www.elsevier.com/locate/jos
Original Article
Short-term mortality and risk factors associated with reoperation after emergency amputation of the lower limbs Shuichi Miyamoto a, *, Makoto Otsuka a, Fumio Hasue a, Takayuki Fujiyoshi a, Koushirou Kamiya a, Hitoshi Kiuchi a, Tadashi Tanaka a, Junichi Nakamura b, Sumihisa Orita c, Seiji Ohtori b a
Kimitsu Central Hospital, 1010 Sakurai, Kisarazu City, Chiba, 292-8535, Japan Graduate School of Medicine, Chiba University, 1-8-1 Inohana, Chuo-ku, Chiba City, Chiba, 260-8677, Japan Center for Advanced Joint Function and Reconstructive Spine Surgery, Graduate School of Medicine, Chiba University, 1-8-1 Inohana, Chuo-ku, Chiba, 2608670, Japan b c
a r t i c l e i n f o
a b s t r a c t
Article history: Received 20 December 2018 Received in revised form 14 May 2019 Accepted 26 May 2019 Available online xxx
Background: As Japan has an aging population, elderly patients with multiple complicating comorbidities have increased in number. The purpose of the present study was to examine the 90-day mortality and reoperation rate after emergency amputation of the lower limbs and to clarify patient characteristics and preoperative comorbidities that are risk factors for reoperations. Methods: This retrospective case study was conducted between January 2005 and December 2017. Ninety-one patients with 108 limbs who had undergone emergency amputation of the lower limbs were included in this study. Baseline data, medical comorbidities, surgical procedures and laboratory data including C-reactive protein (CRP), white blood cells (WBC) and hemoglobin A1c (HbA1c) were evaluated. Outcome measurements were mortality, reoperation rate, and factors associated with reoperation in 90 days and any complication after amputation of the lower limbs. Results: The patients (male/female ratio ¼ 64/27) had a mean age of 64.9 years, a mean WBC count of 16,747/mL, a mean CRP value of 15.6 mg/dL, and a mean HbA1c value of 7.5%. The 90-day mortality rate in the study population was 12.1%. The 90-day reoperation rate was 35.2% (32/91) patients with 32.4% (35/ 108) limbs. The factor significantly associated with reoperation in 90 days was peripheral arterial disease (PAD) (p ¼ 0.022, HR ¼ 2.75), identified with Cox proportional hazards regression analysis. A total of 41.8% (38/91) patients with 40.7% (44/108) limbs experienced postoperative wound complications. The overall rate of medical complications was 18.7% (17/91) patients with 16.7% (18/108) limbs. Conclusions: This study indicates that the 90-day mortality and reoperation rates after emergency amputation of the lower limbs were high. Furthermore, PAD influences the rate of reoperation in 90 days. However, there was no association between laboratory results or diabetes mellitus and reoperation. These findings will have important implications for the treatment of lower leg amputation and prevention of reoperation. © 2019 The Japanese Orthopaedic Association. Published by Elsevier B.V. All rights reserved.
1. Introduction
* Corresponding author. Department of Orthopaedic Surgery, Kimitsu Central Hospital, 1010 Sakurai, Kisarazu City, Chiba, 292-8535, Japan. Fax: þ81 43 836 3867. E-mail addresses: [email protected] (S. Miyamoto), [email protected] (M. Otsuka), [email protected] (F. Hasue), [email protected] (T. Fujiyoshi), [email protected] (K. Kamiya), [email protected] (H. Kiuchi), [email protected] (T. Tanaka), [email protected] (J. Nakamura), [email protected] (S. Orita), sohtori@faculty. chiba-u.jp (S. Ohtori).
Japan has an aging society, in which 27.3% of the population was equal to or over 65 years old in 2016 according to the Japanese bureau of statistics. In this situation, the most common disorders are hypertension and diabetes mellitus, according to a 2014 survey by the Health Ministry, in 10.1 million and 3.16 million people respectively. Furthermore, it is well known that either of these conditions may coexist with peripheral arterial disease (PAD) [1,2]. Endovascular revascularization techniques
https://doi.org/10.1016/j.jos.2019.05.018 0949-2658/© 2019 The Japanese Orthopaedic Association. Published by Elsevier B.V. All rights reserved.
Please cite this article as: Miyamoto S et al., Short-term mortality and risk factors associated with reoperation after emergency amputation of the lower limbs, Journal of Orthopaedic Science, https://doi.org/10.1016/j.jos.2019.05.018
2
S. Miyamoto et al. / Journal of Orthopaedic Science xxx (xxxx) xxx
and bypass surgery have greatly improved patients’ quality of life. However, some elderly patients have multiple complicating comorbidities, so that in some cases the blood circulation of lower limbs cannot be salvaged with any interventions. In other words, emergency amputation of the lower limbs is often performed to save a life by not only orthopedic trauma surgeons, but also general surgeons. It is well recognized that the combination of multiple comorbidities, poor perioperative general condition and invasiveness of surgery is associated with higher risk of mortality. Several studies have shown severe morbidities, complications and reoperations after surgery for amputation of the lower limbs [3e6]. Furthermore, postoperative complications that include medical complications lead to worsened general condition and increased hospital and patient expenditures. Evaluating the factors associated with complications after amputation may be an important advance in the design of strategies. The aim of this study was to investigate the 90-day mortality and reoperation rates after emergency amputation of the lower limbs and to clarify patient characteristics and preoperative comorbidities that are risk factors for reoperation within 90 days. 2. Materials and methods 2.1. Study design and data source This retrospective study was conducted between January 2005 and December 2017 in Kimitsu Central Hospital (Kisarazu City, Chiba, Japan). All procedures performed in study involving human participants were in accordance with the ethical standards approval of the institutional review board and with the 1964 Helsinki declaration and its later amendments or comparable ethical standards. 2.2. Patients The patients included in the study had undergone emergency amputation of the lower limbs for the purpose of lifesaving during this period and were followed for a minimum of 90 days after the operation (Fig. 1). In referrals from other institutions and in consultations from our institution, emergency amputation was indicated for sepsis caused by foot ulcers with infection in patients with diabetes, and for sepsis caused by combined critical limb ischemia and infection in patients with peripheral arterial
disease or diabetes. For trauma and hypovolemic shock, amputation was indicated for patients with mangled extremity severity score (MESS) 7 points [7]. Emergency amputation was defined as an amputation that was performed immediately at the time of emergency transportation in a patient with trauma, or performed within 24 h after consultation with the emergency physician or medical physician in patients with sepsis caused by limb ischemia and infection. Reasons for consultation included noncontrollable sepsis. Patients who had undergone amputation below the ankle were excluded, such as Syme, Chopart, Lisfranc, transmetatarsal, metatarsal phalangeal disarticulation and toe amputation [8]. Patients' baseline data and comorbidities were analyzed from their medical records retrospectively. Baseline data included age, sex, body mass index (BMI) at the time of operation, American Society of Anesthesiologists (ASA) grade [9,10], amputation level and the presence of a traumatic event. Patient age was categorized into groups: <60, 60e79, and 80 years. BMI was categorized into groups: <25 and 25 kg/m2. ASA grade was classified as mild (grade 1e2) or severe (grade 3e5). The amputation level was classified as above-knee (AK) and below-knee (BK). Comorbidities included a history of hypertension, diabetes mellitus, PAD, liver disease, dialysis, heart failure, coronary artery disease (CAD), and smoking. Hypertension and diabetes mellitus were defined as the use of antihypertensive or antidiabetic medications. PAD was diagnosed by computed tomography angiography (CTA), taking into account the patient's renal dysfunction. If CTA evaluation could not be performed due to the patient's condition, PAD was diagnosed based on the patient's medical diagnosis history from our institution or others. Other comorbidities were defined based on medical records of past or current treatment in private practice or at our hospital. Liver disease included chronic hepatitis B virus, hepatitis C virus and hepatic cirrhosis. Current smoking was defined as any amount of tobacco use, including abstinence, within 1 year before the operation. The type of anesthesia was analyzed and classified as general, spinal, and block. Individual anesthesiologists determined the ASA score and the type of anesthesia. The following relevant laboratory data were investigated just before surgery: C-reactive protein (CRP), white blood cells (WBC) and HbA1c. CRP was categorized into groups: <10, 10e19, or 20 mg/ dL. WBC was categorized into groups: <9000, 9000e20000, or 20,000/mL. HbA1c was categorized into groups: <6.5, 6.5e9, or 10 mg/dL.
Fig. 1. Flow diagram for the study.
Please cite this article as: Miyamoto S et al., Short-term mortality and risk factors associated with reoperation after emergency amputation of the lower limbs, Journal of Orthopaedic Science, https://doi.org/10.1016/j.jos.2019.05.018
S. Miyamoto et al. / Journal of Orthopaedic Science xxx (xxxx) xxx
2.2.1. Operative procedure Depending on the renal function of the patients, the feasibility of CTA to evaluate peripheral circulation was considered. In the patients that could be evaluated by CTA, the amputation level was based on the level of disrupted blood circulation as shown by the imaging. If CTA was not feasible, or if the circulation was not disrupted in patients with diabetes and sepsis, the amputation level was decided on the basis of physical findings including reddening, swelling and necrosis. BK amputation was indicated for the area of ischemia and infection from the foot to the distal third of the crus. AK amputation was indicated if proximal from the distal third of the crus. In the patient with trauma and hypovolemic shock, amputation level was based on physical findings including bone, muscle and arterial injury. All patients were placed in the supine position on a standard table. The affected limbs were cleaned and draped with a tourniquet. AK and BK amputation was performed in accordance with surgical technique [8]. Amputation level was short or middle transfemoral in AK and 8.8e12.5 cm distal to the line of the knee joint in BK [8]. The drainage tube was inserted in the area of osteotomy and removed 1 or 2 days postoperatively. 2.3. All-cause mortality in 90 days after surgery The medical records were obtained and examined to calculate all-cause mortality at 90 days postoperatively. 2.4. Reoperations and any complication in 90 days Unplanned reoperation and any postoperative complications noted in the medical records were included. Unplanned reoperation was defined as surgical debridement, re-suture and additional amputation. Complications were categorized as wound and medical complications. The latter included diagnosis of pneumonia, urinary tract infection, sepsis, heart failure, renal failure requiring dialysis, and other conditions (e.g., pleural effusion and hypoalbuminemia). Hypoalbuminemia was defined as a level of serum albumin below 2.0 g/dl. 2.5. Statistical analysis All variables were summarized as frequencies, percentages, means, and standard deviations. Using the KolmogoroveSmirnov test, normality tests were performed. Univariate correlates for baseline data and comorbidities were tested using the Fisher exact probability test for dichotomous data and the ManneWhitney U test for continuous variables. P values < 0.05 were considered significant. Reoperation was a dichotomous dependent variable. Univariate Cox proportional hazards regression analyses were used to determine the relationship between reoperation and and tertiles of age (<60, 60e79, or 80 years), sex (male/female), BMI (<25/ 25 kg/m2), ASA (1e2/3e5), level of amputation (AK/BK), presence of traumatic event, comorbidity (hypertension, diabetes mellitus, PAD, liver disease, dialysis, arrhythmia, heart failure, CAD and smoking), tertiles of CRP (<10, 10e19, or 20 mg/dL), tertiles of WBC (<9000, 9000e20000, or 20,000/mL) and tertiles of HbA1c (<6.5, 6.5e9, or 10 mg/dL). All factors with a P value of <0.1 in the univariate analysis were used in multivariate Cox proportional hazards regression analysis. In the multivariate analysis, a value of P < 0.05 was considered significant and was defined to identify all influencing factors for reoperations. All results are expressed as the P value, hazard ratio (HR), and corresponding confidence interval. All calculations were performed using R software, version 3.4.1 (R Development Core Team).
3
3. Results 3.1. Patients A total of 143 patients with 169 limbs had undergone emergency amputation of the lower limbs to save their lives during the study period. Among these patients, 91 patients with 108 limbs were followed for a minimum of 90 days after their operations. The baseline characteristics of the number of patients with limbs enrolled are summarized in Table 1. The characteristics of the patients in the groups with diabetes mellitus, PAD, or both are outlined in Table 2. 3.2. Mortality in 90 days The 90-day mortality rate in the study population was 12.1% (11/ 91) using KaplaneMeier estimates in Fig. 2, with male mortality at 14.1% (9/64) and female mortality at 7.4% (2/27). The mean age of these patients at death was 73.8 ± 15.4 years [range, 31e87 years]. The average survival period after operation was 36.3 ± 32.6 days [range, 0e87 days]. All patients had severe ASA grade (grade 3e5). A total of 81.8% (9/11) patients underwent AK amputation, and 90.9% (10/11) patients had PAD or diabetes mellitus; 27.3% (3/11) died within 1 week after the operation. In the remainder, who
Table 1 Characteristics of the study population (91 patients with 108 limbs). Variable
Total number of patients/limbs, n (%)
Age (years)a Sex Male Female BMI (kg/m2)a ASA gradea Amputation level AK Total Unilateral Bilateral BK Total Unilateral Bilateral AK after BK Total Unilateral Bilateral Traumatic event Medical comorbidities Hypertension Diabetes mellitus Peripheral arterial disease Liver disease Dialysis Arrhythmia Heart failure Coronary artery disease Smoking Anesthesia General General þ Spinal General þ Block Spinal Block CRP (mg/dL)a WBC (/mL)a HbA1c-NGSP (%)a Missing data
64.9 ± 15.5 64 (70.3)/76 (70.4) 27 (29.7)/32 (29.6) 22.1 ± 4.2 2.84 ± 0.7
57 (62.6)/61 (56.5) 53 (93.0)/53 (86.9) 4 (7.0)/8 (13.1) 42 (46.2)/47 (43.5) 37 (88.1)/37 (78.7) 5 (11.9)/10 (21.3) 8 (8.8)/8 (7.4) 8 (100)/8 (100) e 15 (16.5)/15 (13.9) 48 (52.7)/59 (54.6) 50 (54.9)/60 (55.6) 45 (49.5)/59 (54.6) 9 (9.9)/10 (9.3) 13 (14.3)/22 (20.4) 15 (16.5)/19 (17.6) 16 (17.6)/21 (19.4) 16 (17.6)/19 (17.6) 33 (36.3)/42 (38.9) 62 (68.1)/73 (67.6) 8 (8.8)/11 (10.2) 9 (9.9)/10 (9.3) 7 (7.7)/7 (6.5) 5 (5.5)/7 (6.5) 15.6 ± 11.2 16,747 ± 9311 7.5 ± 2.5 8 (7.7)/8 (7.4)
CRP: C-reactive protein, HbA1c-NGSP: HbA1c-National Glycohemoglobin Standardization Program, WBC: white blood cell count. a The values are given as the mean and standard deviation.
Please cite this article as: Miyamoto S et al., Short-term mortality and risk factors associated with reoperation after emergency amputation of the lower limbs, Journal of Orthopaedic Science, https://doi.org/10.1016/j.jos.2019.05.018
4
S. Miyamoto et al. / Journal of Orthopaedic Science xxx (xxxx) xxx
Table 2 Characteristics of the diabetes mellitus and peripheral arterial disease patients (76 patients with 93 limbs). Variable
Frequency (%) 76 (100)/93 (100)
Number of patients/limbs Age (years)d Sex Male Female BMI (kg/m2)d ASA graded Amputation level AK BK Medical comorbidities Hypertension Liver disease Dialysis Arrhythmia Heart failure Coronary artery disease Smoking CRP (mg/dL)d WBC (/mL)d HbA1c-NGSP (%)d Disruption level of blood flow Aortoiliac Femoropopliteal Infrapopliteal Lack of CTA data
Diabetes mellitus
Peripheral arterial disease
Combined diabetes mellitus and peripheral arterial disease
31 (40.8)/34 (36.6) 62.6 ± 12.0
26 (34.2)/33 (35.5) 72.3 ± 13.7a
19 (25.0)/26 (28.0) 67.8 ± 8.8c
19 (61.3)/22 (64.7) 12 (38.7)/12 (35.3) 23.3 ± 4.3 2.9 ± 0.5
19 (73.1)/24 (72.7) 7 (26.9)/9 (27.3) 19.7 ± 3.5a 2.9 ± 0.6
14 (73.7)/18 (69.2) 5 (26.3)/8 (30.8) 22.1 ± 3.5c 3.0 ± 0.2
15 (48.4)/15 (44.1) 18 (58.1)/19 (55.9)
24 (92.3)/27 (81.8)a 5 (19.2)/6 (18.2)a
10 (52.6)/11 (42.3)c 12 (63.2)/15 (57.7)c
15 (48.4)/18 (52.9) 2 (6.5)/2 (5.9) 1 (3.2)/2 (5.9) 1 (3.2)/1 (2.9) 3 (9.7)/3 (8.8) 2 (6.5)/2 (5.9) 9 (29.0)/11 (32.4) 19.6 ± 12.3 18,212 ± 10,168 9.3 ± 2.8
16 (61.5)/20 (60.6) 3 (11.5)/4 (12.1) 7 (26.9)/11 (33.3)a 10 (38.5)/11 (33.3)a 6 (23.1)/7 (21.2) 8 (30.8)/9 (27.3)a 10 (38.5)/13 (39.4) 13.2 ± 9.9a 15,958 ± 9201 5.5 ± 0.5a
15 (78.9)/19 (73.1)b 2 (10.5)/2 (7.7) 5 (26.3)/9 (34.6)b 4 (21.1)/7 (26.9) 7 (36.8)/11 (42.3)b 6 (31.6)/8 (30.8)b 9 (47.4)/13 (50.0) 15.0 ± 9.4 16,600 ± 9764 8.2 ± 1.7c
e e e e
4 (15.4)/4 (12.1) 16 (61.5)/20 (60.6) 2 (7.7)/2 (6.1) 5 (19.2)/7 (21.2)
1 8 6 6
(5.3)/1 (3.8) (42.1)/11 (42.3) (31.6)/7 (26.9) (31.6)/7 (26.9)
CRP: C-reactive protein, HbA1c-NGSP: HbA1c-National Glycohemoglobin, WBC: white blood cell count, CTA: computed tomography angiography. a Significant difference between diabetes mellitus and peripheral arterial disease (p < 0.05). b Significant difference between diabetes mellitus and combined diabetes mellitus and peripheral arterial disease (p < 0.05). c Significant difference between peripheral arterial disease and combined diabetes mellitus and peripheral arterial disease (p < 0.05). d The values are given as the mean and standard deviation.
Of the 91 patients with 108 limbs, 41.8% (38/91) patients with 40.7% (44/108) limbs experienced postoperative wound complications within 90 days. The overall rate of medical complications within 90 days was 18.7% (17/91) patients, which included pneumonia and urinary tract infection in 6.6% (6/91) patients, sepsis in 4.4% (4/91) patients, heart failure in 1.1% (1/91) patients, renal failure requiring dialysis in 1.1% (1/91) patients, and other complications in 4.4% (4/91) patients. 4. Discussion
Fig. 2. KaplaneMeier survivorship curve with mortality and reoperation as the end point.
survived longer, the average survival period after the operation was 52.1 ± 28.2 days [range, 8e87 days].
3.3. Reoperations and any complication in 90 days The 90-day reoperation rate was 35.2% (32/91) patients with 32.4% (35/108) limbs. Survival analysis with reoperation as the end point is shown in Fig. 2. Univariate Cox regression analyses of 10 parameters for reoperation are summarized in Table 3. All factors with a p value of <0.1 in the univariate analysis were used in multivariate analysis (Table 3). Multivariate analysis revealed that reoperation within 90 days after emergency amputation of the lower limbs was significantly associated with PAD (p ¼ 0.022, HR ¼ 2.75) (Table 4).
To our knowledge, there have been only a few reports about postoperative mortality and reoperation rates after emergency amputation of the lower limbs. Contrary to expectation, the 90-day mortality and reoperation rates after emergency amputation of the lower limbs were high, with a mortality rate of 12.1% and a reoperation rate of 35.2% of patients with 32.4% of limbs. We also found that patients with PAD had a significantly higher risk of reoperation after emergency amputation of the lower limbs. No association was found between laboratory data such as CRP, WBC and HbA1c and reoperation, and furthermore, between diabetes mellitus and reoperation. Also, 41.8% of patients with 40.7% of limbs experienced wound complications and 18.7% of patients with 16.7% of limbs suffered medical complications postoperatively. Since the recent development of endovascular revascularization techniques and bypass surgery, blood circulation of the lower legs was increased in critical limb ischemia and severe trauma. Furthermore, lower leg function and quality of life were greatly improved. However, circulation in the lower limbs could not be salvaged in some elderly patients with multiple complicating comorbidities. In general, the principal objective of amputation is
Please cite this article as: Miyamoto S et al., Short-term mortality and risk factors associated with reoperation after emergency amputation of the lower limbs, Journal of Orthopaedic Science, https://doi.org/10.1016/j.jos.2019.05.018
S. Miyamoto et al. / Journal of Orthopaedic Science xxx (xxxx) xxx
5
Table 3 Univariate analyses of 10 parameters for reoperations (n ¼ 91, p < 0.1). Variable
Number of patients/limbs Age category (years) <60 60e79 80 Sex Male Female BMI (kg/m2) <25 25 ASA grade 1e2 3e5 Amputation level AK BK Traumatic event Medical comorbidities Hypertension Diabetes mellitus Peripheral arterial disease Liver disease Dialysis Arrhythmia Heart failure Coronary artery disease Smoking CRP (mg/dL) <10 10e19 20 Unspecified WBC (/mL) <9000 9000e20000 20,000 HbA1c-NGSP (%) <6.5 6.5e9 10 Unspecified
Frequency (%)
HR (95% CI)
p value
7 (21.9)/8 (22.9) 18 (56.2)/20 (57.1) 7 (21.9)/7 (20.0)
1.0 1.45 (0.64e3.29) 2.06 (0.75e5.68)
e 0.376 0.164
43 (72.9)/53 (72.6) 16 (27.1)/20 (27.4)
21 (65.6)/23 (65.7) 11 (34.4)/12 (34.3)
1.0 1.41 (0.70e2.83)
e 0.336
77 (84.6)/92 (85.2) 14 (15.4)/16 (14.8)
49 (83.1)/61 (83.6) 10 (16.9)/12 (16.4)
28 (87.5)/31 (88.6) 4 (12.5)/4 (11.4)
1.0 0.69 (0.24e1.95)
e 0.48
22 (24.2)/23 (21.3) 69 (75.8)/85 (78.7)
15 (25.4)/15 (20.5) 44 (74.6)/58 (79.5)
7 (21.9)/8 (22.9) 25 (78.1)/27(77.1)
1.0 0.99 (0.45e2.18)
e 0.98
57 (62.6)/61 (56.5) 42 (46.2)/47 (43.5) 15 (16.5)/15 (13.9)
41 (69.5)/43 (58.9) 25 (42.4)/30 (41.1) 8 (13.6)/8 (11.0)
16 (0.5)/18 (51.4) 17 (53.1)/17 (48.6) 7(21.9)/7(20.0)
1.0 1.13 (0.58e2.20) 1.74 (0.76e3.98)
e 0.71 0.192
48 (52.7)/59 (54.6) 50 (54.9)/60 (55.6) 45 (49.5)/59 (54.6) 9 (9.9)/10 (9.3) 13 (14.3)/22 (20.4) 15 (16.5)/19 (17.6) 16 (17.6)/21 (19.4) 16 (17.6)/19 (17.6) 33 (36.3)/42 (38.9)
26 (44.1)/34 (46.6) 34 (57.6)/43 (58.9) 21 (35.6)/32 (43.8) 7 (11.9)/8 (11.0) 6 (10.2)/14(19.2) 6 (10.2)/10 (13.7) 7 (11.9)/11 (15.1) 8 (13.6)/10 (13.7) 23 (39.0)/31 (42.5)
22 (68.8)/25 (71.4) 16 (50.0)/17 (48.6) 24 (75.0)/27 (77.1) 2 (6.3)/2 (5.7) 7 (21.9)/8 (22.9) 9 (28.1)/9 (25.7) 9 (28.1)/10 (28.6) 8 (25.0)/9 (25.7) 10 (31.3)/11(31.4)
2.03 0.62 3.43 0.52 1.12 1.80 2.04 1.84 0.67
(0.97e4.23) (0.32e1.21) (1.56e7.56) (0.12e2.16) (0.51e2.46) (0.84e3.85) (0.98e4.25) (0.86e3.94) (0.33e1.36)
0.059 0.163 0.022 0.365 0.782 0.13 0.058 0.114 0.267
29 (31.9)/40 (37.0) 29 (31.9)/32 (29.6) 32 (35.2)/35 (32.4) 1 (1.1)/1 (0.9)
16 (27.1)/25(34.2) 19 (32.2)/20(27.4) 24 (40.7)/28 (38.4)
14 (43.8)/15 (42.9) 10 (31.3)/12 (34.3) 7 (21.9)/7 (20.0)
1.0 1.03 (0.48e2.20) 0.54 (0.22e1.32)
e 0.941 0.173
12 (13.2)/18 (16.7) 56 (61.5)/62 (57.4) 23 (25.3)/28 (25.9)
8 (13.6)/12(16.4) 34 (57.6)/39(53.4) 17 (28.8)/22(30.1)
4 (12.5)/6 (17.1) 22 (68.7)/23 (65.7) 6 (18.8)/6(17.1)
1.0 1.26 (0.51e3.11) 0.67 (0.22e2.09)
e 0.61 0.494
40 (44.0)/48 (44.4) 30 (33.0)/39 (36.1) 13 (14.2)/13 (12.0) 8 (8.8)/8 (7.4)
26 (44.1)/32 (43.8) 18 (30.5)/26 (35.6) 11 (18.6)/11 (15.1)
14 (43.8)/16 (45.7) 12 (37.5)/13 (37.1) 2 (6.3)/2 (5.7)
1.0 0.86 (0.42e1.80) 0.36 (0.08e1.57)
e 0.696 0.174
Total
No reoperation
Reoperation
91 (100)/108 (100)
59 (64.8)/73 (67.6)
32 (35.2)/35 (32.4)
27 (29.7)/31 (28.7) 46 (50.5)/57 (52.8) 18 (19.8)/20 (18.5)
20 (33.9)/23 (31.5) 28 (47.5)/37 (50.7) 11 (18.6)/13 (17.8)
64 (70.3)/76 (70.4) 27 (29.7)/32 (29.6)
Categorical variables are presented as number (%). Bold p-values are statistically significant. HR: hazard ratio, CI: confidence interval; BMI: body mass index, ASA: American Society of Anesthesiology, AK: above the knee, BK: below the knee, CRP: C-reactive protein, HbA1c-NGSP: HbA1c-National Glycohemoglobin Standardization Program, WBC: white blood cell count.
Table 4 Multivariate analyses of 3 parameters for reoperations (n ¼ 91, p < 0.05). Variable
HR (95%CI)
p value
Hypertension Peripheral arterial disease Heart failure
1.41 (0.64e3.09) 2.75 (1.16e6.53) 1.51 (0.70e3.24)
0.392 0.022 0.289
Bold p-values are statistically significant. HR: hazard ratio, CI: confidence interval.
to improve the pain caused by ischemia, to control the spreading area of infection and gangrene, and to treat septic and hypovolemic shock. A strong association between major amputation and certain comorbidities has been reported. Patients with diabetes mellitus were more likely to need a major amputation, compared to nondiabetic patients with PAD [11,12]. Diabetes mellitus, chronic obstructive pulmonary disease and systemic vasculitis were also associated with a significant risk of lower extremity amputation in severe peripheral arterial occlusive disease [13]. According to several reports, mortality rates after amputation were considerable.
The overall mortality rates after major amputation as described by Pell J et al. were 29% at 6 months, 37% at 1 year, and 49% at 2 years [6]. According to a retrospective study by Klaphake S, the overall mortality rates of patients aged >70 years who were treated for chronic critical limb ischemia by primary or secondary major amputation were 30 days, 14%; 6 months, 38%; 1 year, 44%; 3 years, 66% and 5 years, 85%, respectively [14]. These reports suggested that the short- and long-term mortality rates of these patients were unpredictably high. Although our examination was over a shorter term, our results concur with the other reports. In our study, all patients had ASA grades in the severe range, and almost all patients underwent AK amputation and had PAD or diabetes mellitus as comorbidities. A review of the literature about re-amputation rates after amputation types including toe, ray, midfoot (transmetatarsal amputation, Lisfranc, or Chopart), and major amputations (Syme, transtibial, transfemoral, or hip disarticulation) reported the following rates: 1 year, 26.7%; 3 years, 48.3% and 5 years; 60.7% [15]. Furthermore, the rates of re-amputation on the ipsilateral limb after a major amputation were: 1 year, 4.7%; 3 year, 11.8%; and 5 year, 13.3% [15]. Although our 90-day reoperation rate of 35.2%
Please cite this article as: Miyamoto S et al., Short-term mortality and risk factors associated with reoperation after emergency amputation of the lower limbs, Journal of Orthopaedic Science, https://doi.org/10.1016/j.jos.2019.05.018
6
S. Miyamoto et al. / Journal of Orthopaedic Science xxx (xxxx) xxx
patients with 32.4% limbs was higher than those of reports, these patients might have had a different indication for reoperation, including surgical debridement and re-suture. Our study also found a significant association between PAD and a higher risk of reoperation. Reportedly, the risk factor for an association between reoperation and amputation of the lower limbs is unknown. Hence, our findings may have important implications for the management of amputation of the lower limbs. Our findings should be interpreted in the context of several limitations. First, this was a nonrandomized retrospective study and the sample size was relatively small. Hence, its power for detecting significance among risk factors for reoperation remains low. Second, our study had a short-term follow-up period. The influence of patient characteristics and preoperative comorbidities on long-term mortality should be taken into consideration in future studies. Third, these results were influenced by selection bias; the indication for amputation and the amputation level were decided by individual physicians, and CTA to determine the amputation level was not performed on all patients because some had renal dysfunction in sepsis. Fourth, we did not evaluate medication or fluid management during surgery, or the volume of the surgeons. Fifth, therapeutic options including medication and endovascular revascularization techniques have increased and progressed over time in this study. Furthermore, therapeutic options were decided by individual medical physicians. Hence, the treatment progress leading to amputation varied across the patients. In conclusion, the results of this retrospective case study indicate that the 90-day mortality and reoperation rate after emergency amputation of the lower limbs were high. We also found that patients with PAD were at a significantly higher risk of reoperation. On the other hand, no association was found between laboratory results (CRP, WBC and HbA1c) or diabetes mellitus and reoperation. This study contributes to the literature, as these findings will have important implications for treatment of lower leg amputation and prevention of reoperation. Furthermore, the study findings can provide a useful management strategy for not only orthopedic trauma surgeons, but also general surgeons. Conflicts of interest The authors declare that there are no conflicts of interest regarding the publication of this paper. Acknowledgements We are grateful to Dr Makoto Otsuka, Dr Fumio Hasue, Dr Takayuki Fujiyoshi, Dr Koushirou Kamiya, Dr Hitoshi Kiuchi, Dr Tadashi Tanaka, Dr Junichi Nakamura, Dr Sumihisa Orita, and Professor Seiji Ohtori, head of our department, for their valuable suggestions and support during the course of this study. All authors
have read and approved the submitted version. Study design: Shuichi Miyamoto, Makoto Otsuka, Tadashi Tanaka, Junichi Nakamura, Sumihisa Orita, Seiji Ohtori. Drafting the protocol: Shuichi Miyamoto, Makoto Otsuka, Tadashi Tanaka, Sumihisa Orita, Seiji Ohtori. Advice on the statistical analysis: Shuichi Miyamoto, Sumihisa Orita. Patient recruitment and data collection: Shuichi Miyamoto, Fumio Hasue, Takayuki Fujiyoshi, Koushirou Kamiya, Hitoshi Kiuchi. First draft of the manuscript: Shuichi Miyamoto. We did not receive any funding or financial support that may be perceived to have biased the study.
References [1] Jude EB, Oyibo SO, Chalmers N, Boulton AJ. Peripheral arterial disease in diabetic and nondiabetic patients: a comparison of severity and outcome. Diabetes Care 2001;24(8):1433e7. [2] Brownrigg JR, Apelqvist J, Bakker K, Schaper NC, Hinchliffe RJ. Evidence-based management of PAD & the diabetic foot. Eur J Vasc Endovasc Surg 2013;45(6): 673e81. [3] Norgren L, Hiatt WR, Dormandy JA, Nehler MR, Harris KA, Fowkes FG. TASC II working group. Inter-Society consensus for the management of peripheral arterial disease (TASC II). J Vasc Surg 2007;45(Suppl S):S5e67. [4] Reinecke H, Unrath M, Freisinger E, Bunzemeier H, Meyborg M, Lüders F, Gebauer K, Roeder N, Berger K, Malyar NM. Peripheral arterial disease and critical limb ischaemia: still poor outcomes and lack of guideline adherence. Eur Heart J 2015;36(15):932e8. [5] Dormandy JA. Natural history of patients with peripheral occlusive arterial disease. Ann Chir Gynaecol 1992;81(2):86e8. [6] Pell J, Stonebridge P. Association between age and survival following major amputation. The Scottish vascular audit group. Eur J Vasc Endovasc Surg 1999;17(2):166e9. [7] Johansen K, Daines M, Howey T, Helfet D, Hansen Jr ST. Objective criteria accurately predict amputation following lower extremity trauma. J Trauma 1990;30(5):568e72. [8] Terry Canale S, James H. In: Beaty Campbell's operative orthopaedics. 12th ed. Philadelphia: Elsevier Mosby; 2013. p. 637e50 [Amputations of the lower extremity]. [9] Saklad M. Grading of patients for surgical procedures. Anesthesiology 1941;2(3):281e4. [10] Dripps RD, Lamont A, Eckenhoff JE. The role of anesthesia in surgical mortality. JAMA 1961;178(3):261e6. [11] Carmona GA, Hoffmeyer P, Herrmann FR, Vaucher J, Tschopp O, Lacraz A, Vischer UM. Major lower limb amputations in the elderly observed over ten years: the role of diabetes and peripheral arterial disease. Diabetes Metab 2005;31(5):449e54. hn AB, Atroshi I. [12] Johannesson A, Larsson GU, Ramstrand N, Turkiewicz A, Wire Incidence of lower-limb amputation in the diabetic and nondiabetic general population: a 10-year population-based cohort study of initial unilateral and contralateral amputations and reamputations. Diabetes Care 2009;32(2): 275e80. [13] Lu MC, Hsu H, Lin CH, Koo M. Systemic vasculitis is associated with a higher risk of lower extremity amputation in patients with severe peripheral arterial occlusive disease: a secondary analysis of a nationwide, population-based health claims database. Rheumatol Int 2017;37(11):1847e52. [14] Klaphake S, de Leur K, Mulder PG, Ho GH, de Groot HG, Veen EJ, Verhagen HJ, van der Laan L. Mortality after major amputation in elderly patients with critical limb ischemia. Clin Interv Aging 2017;12:1985e92. [15] Izumi Y, Satterfield K, Lee S, Harkless LB. Risk of reamputation in diabetic patients stratified by limb and level of amputation: a 10-year observation. Diabetes Care 2006;29(3):566e70.
Please cite this article as: Miyamoto S et al., Short-term mortality and risk factors associated with reoperation after emergency amputation of the lower limbs, Journal of Orthopaedic Science, https://doi.org/10.1016/j.jos.2019.05.018
Contents lists available at ScienceDirect
Journal of Orthopaedic Science journal homepage: http://www.elsevier.com/locate/jos
Original Article
Short-term mortality and risk factors associated with reoperation after emergency amputation of the lower limbs Shuichi Miyamoto a, *, Makoto Otsuka a, Fumio Hasue a, Takayuki Fujiyoshi a, Koushirou Kamiya a, Hitoshi Kiuchi a, Tadashi Tanaka a, Junichi Nakamura b, Sumihisa Orita c, Seiji Ohtori b a
Kimitsu Central Hospital, 1010 Sakurai, Kisarazu City, Chiba, 292-8535, Japan Graduate School of Medicine, Chiba University, 1-8-1 Inohana, Chuo-ku, Chiba City, Chiba, 260-8677, Japan Center for Advanced Joint Function and Reconstructive Spine Surgery, Graduate School of Medicine, Chiba University, 1-8-1 Inohana, Chuo-ku, Chiba, 2608670, Japan b c
a r t i c l e i n f o
a b s t r a c t
Article history: Received 20 December 2018 Received in revised form 14 May 2019 Accepted 26 May 2019 Available online xxx
Background: As Japan has an aging population, elderly patients with multiple complicating comorbidities have increased in number. The purpose of the present study was to examine the 90-day mortality and reoperation rate after emergency amputation of the lower limbs and to clarify patient characteristics and preoperative comorbidities that are risk factors for reoperations. Methods: This retrospective case study was conducted between January 2005 and December 2017. Ninety-one patients with 108 limbs who had undergone emergency amputation of the lower limbs were included in this study. Baseline data, medical comorbidities, surgical procedures and laboratory data including C-reactive protein (CRP), white blood cells (WBC) and hemoglobin A1c (HbA1c) were evaluated. Outcome measurements were mortality, reoperation rate, and factors associated with reoperation in 90 days and any complication after amputation of the lower limbs. Results: The patients (male/female ratio ¼ 64/27) had a mean age of 64.9 years, a mean WBC count of 16,747/mL, a mean CRP value of 15.6 mg/dL, and a mean HbA1c value of 7.5%. The 90-day mortality rate in the study population was 12.1%. The 90-day reoperation rate was 35.2% (32/91) patients with 32.4% (35/ 108) limbs. The factor significantly associated with reoperation in 90 days was peripheral arterial disease (PAD) (p ¼ 0.022, HR ¼ 2.75), identified with Cox proportional hazards regression analysis. A total of 41.8% (38/91) patients with 40.7% (44/108) limbs experienced postoperative wound complications. The overall rate of medical complications was 18.7% (17/91) patients with 16.7% (18/108) limbs. Conclusions: This study indicates that the 90-day mortality and reoperation rates after emergency amputation of the lower limbs were high. Furthermore, PAD influences the rate of reoperation in 90 days. However, there was no association between laboratory results or diabetes mellitus and reoperation. These findings will have important implications for the treatment of lower leg amputation and prevention of reoperation. © 2019 The Japanese Orthopaedic Association. Published by Elsevier B.V. All rights reserved.
1. Introduction
* Corresponding author. Department of Orthopaedic Surgery, Kimitsu Central Hospital, 1010 Sakurai, Kisarazu City, Chiba, 292-8535, Japan. Fax: þ81 43 836 3867. E-mail addresses: [email protected] (S. Miyamoto), [email protected] (M. Otsuka), [email protected] (F. Hasue), [email protected] (T. Fujiyoshi), [email protected] (K. Kamiya), [email protected] (H. Kiuchi), [email protected] (T. Tanaka), [email protected] (J. Nakamura), [email protected] (S. Orita), sohtori@faculty. chiba-u.jp (S. Ohtori).
Japan has an aging society, in which 27.3% of the population was equal to or over 65 years old in 2016 according to the Japanese bureau of statistics. In this situation, the most common disorders are hypertension and diabetes mellitus, according to a 2014 survey by the Health Ministry, in 10.1 million and 3.16 million people respectively. Furthermore, it is well known that either of these conditions may coexist with peripheral arterial disease (PAD) [1,2]. Endovascular revascularization techniques
https://doi.org/10.1016/j.jos.2019.05.018 0949-2658/© 2019 The Japanese Orthopaedic Association. Published by Elsevier B.V. All rights reserved.
Please cite this article as: Miyamoto S et al., Short-term mortality and risk factors associated with reoperation after emergency amputation of the lower limbs, Journal of Orthopaedic Science, https://doi.org/10.1016/j.jos.2019.05.018
2
S. Miyamoto et al. / Journal of Orthopaedic Science xxx (xxxx) xxx
and bypass surgery have greatly improved patients’ quality of life. However, some elderly patients have multiple complicating comorbidities, so that in some cases the blood circulation of lower limbs cannot be salvaged with any interventions. In other words, emergency amputation of the lower limbs is often performed to save a life by not only orthopedic trauma surgeons, but also general surgeons. It is well recognized that the combination of multiple comorbidities, poor perioperative general condition and invasiveness of surgery is associated with higher risk of mortality. Several studies have shown severe morbidities, complications and reoperations after surgery for amputation of the lower limbs [3e6]. Furthermore, postoperative complications that include medical complications lead to worsened general condition and increased hospital and patient expenditures. Evaluating the factors associated with complications after amputation may be an important advance in the design of strategies. The aim of this study was to investigate the 90-day mortality and reoperation rates after emergency amputation of the lower limbs and to clarify patient characteristics and preoperative comorbidities that are risk factors for reoperation within 90 days. 2. Materials and methods 2.1. Study design and data source This retrospective study was conducted between January 2005 and December 2017 in Kimitsu Central Hospital (Kisarazu City, Chiba, Japan). All procedures performed in study involving human participants were in accordance with the ethical standards approval of the institutional review board and with the 1964 Helsinki declaration and its later amendments or comparable ethical standards. 2.2. Patients The patients included in the study had undergone emergency amputation of the lower limbs for the purpose of lifesaving during this period and were followed for a minimum of 90 days after the operation (Fig. 1). In referrals from other institutions and in consultations from our institution, emergency amputation was indicated for sepsis caused by foot ulcers with infection in patients with diabetes, and for sepsis caused by combined critical limb ischemia and infection in patients with peripheral arterial
disease or diabetes. For trauma and hypovolemic shock, amputation was indicated for patients with mangled extremity severity score (MESS) 7 points [7]. Emergency amputation was defined as an amputation that was performed immediately at the time of emergency transportation in a patient with trauma, or performed within 24 h after consultation with the emergency physician or medical physician in patients with sepsis caused by limb ischemia and infection. Reasons for consultation included noncontrollable sepsis. Patients who had undergone amputation below the ankle were excluded, such as Syme, Chopart, Lisfranc, transmetatarsal, metatarsal phalangeal disarticulation and toe amputation [8]. Patients' baseline data and comorbidities were analyzed from their medical records retrospectively. Baseline data included age, sex, body mass index (BMI) at the time of operation, American Society of Anesthesiologists (ASA) grade [9,10], amputation level and the presence of a traumatic event. Patient age was categorized into groups: <60, 60e79, and 80 years. BMI was categorized into groups: <25 and 25 kg/m2. ASA grade was classified as mild (grade 1e2) or severe (grade 3e5). The amputation level was classified as above-knee (AK) and below-knee (BK). Comorbidities included a history of hypertension, diabetes mellitus, PAD, liver disease, dialysis, heart failure, coronary artery disease (CAD), and smoking. Hypertension and diabetes mellitus were defined as the use of antihypertensive or antidiabetic medications. PAD was diagnosed by computed tomography angiography (CTA), taking into account the patient's renal dysfunction. If CTA evaluation could not be performed due to the patient's condition, PAD was diagnosed based on the patient's medical diagnosis history from our institution or others. Other comorbidities were defined based on medical records of past or current treatment in private practice or at our hospital. Liver disease included chronic hepatitis B virus, hepatitis C virus and hepatic cirrhosis. Current smoking was defined as any amount of tobacco use, including abstinence, within 1 year before the operation. The type of anesthesia was analyzed and classified as general, spinal, and block. Individual anesthesiologists determined the ASA score and the type of anesthesia. The following relevant laboratory data were investigated just before surgery: C-reactive protein (CRP), white blood cells (WBC) and HbA1c. CRP was categorized into groups: <10, 10e19, or 20 mg/ dL. WBC was categorized into groups: <9000, 9000e20000, or 20,000/mL. HbA1c was categorized into groups: <6.5, 6.5e9, or 10 mg/dL.
Fig. 1. Flow diagram for the study.
Please cite this article as: Miyamoto S et al., Short-term mortality and risk factors associated with reoperation after emergency amputation of the lower limbs, Journal of Orthopaedic Science, https://doi.org/10.1016/j.jos.2019.05.018
S. Miyamoto et al. / Journal of Orthopaedic Science xxx (xxxx) xxx
2.2.1. Operative procedure Depending on the renal function of the patients, the feasibility of CTA to evaluate peripheral circulation was considered. In the patients that could be evaluated by CTA, the amputation level was based on the level of disrupted blood circulation as shown by the imaging. If CTA was not feasible, or if the circulation was not disrupted in patients with diabetes and sepsis, the amputation level was decided on the basis of physical findings including reddening, swelling and necrosis. BK amputation was indicated for the area of ischemia and infection from the foot to the distal third of the crus. AK amputation was indicated if proximal from the distal third of the crus. In the patient with trauma and hypovolemic shock, amputation level was based on physical findings including bone, muscle and arterial injury. All patients were placed in the supine position on a standard table. The affected limbs were cleaned and draped with a tourniquet. AK and BK amputation was performed in accordance with surgical technique [8]. Amputation level was short or middle transfemoral in AK and 8.8e12.5 cm distal to the line of the knee joint in BK [8]. The drainage tube was inserted in the area of osteotomy and removed 1 or 2 days postoperatively. 2.3. All-cause mortality in 90 days after surgery The medical records were obtained and examined to calculate all-cause mortality at 90 days postoperatively. 2.4. Reoperations and any complication in 90 days Unplanned reoperation and any postoperative complications noted in the medical records were included. Unplanned reoperation was defined as surgical debridement, re-suture and additional amputation. Complications were categorized as wound and medical complications. The latter included diagnosis of pneumonia, urinary tract infection, sepsis, heart failure, renal failure requiring dialysis, and other conditions (e.g., pleural effusion and hypoalbuminemia). Hypoalbuminemia was defined as a level of serum albumin below 2.0 g/dl. 2.5. Statistical analysis All variables were summarized as frequencies, percentages, means, and standard deviations. Using the KolmogoroveSmirnov test, normality tests were performed. Univariate correlates for baseline data and comorbidities were tested using the Fisher exact probability test for dichotomous data and the ManneWhitney U test for continuous variables. P values < 0.05 were considered significant. Reoperation was a dichotomous dependent variable. Univariate Cox proportional hazards regression analyses were used to determine the relationship between reoperation and and tertiles of age (<60, 60e79, or 80 years), sex (male/female), BMI (<25/ 25 kg/m2), ASA (1e2/3e5), level of amputation (AK/BK), presence of traumatic event, comorbidity (hypertension, diabetes mellitus, PAD, liver disease, dialysis, arrhythmia, heart failure, CAD and smoking), tertiles of CRP (<10, 10e19, or 20 mg/dL), tertiles of WBC (<9000, 9000e20000, or 20,000/mL) and tertiles of HbA1c (<6.5, 6.5e9, or 10 mg/dL). All factors with a P value of <0.1 in the univariate analysis were used in multivariate Cox proportional hazards regression analysis. In the multivariate analysis, a value of P < 0.05 was considered significant and was defined to identify all influencing factors for reoperations. All results are expressed as the P value, hazard ratio (HR), and corresponding confidence interval. All calculations were performed using R software, version 3.4.1 (R Development Core Team).
3
3. Results 3.1. Patients A total of 143 patients with 169 limbs had undergone emergency amputation of the lower limbs to save their lives during the study period. Among these patients, 91 patients with 108 limbs were followed for a minimum of 90 days after their operations. The baseline characteristics of the number of patients with limbs enrolled are summarized in Table 1. The characteristics of the patients in the groups with diabetes mellitus, PAD, or both are outlined in Table 2. 3.2. Mortality in 90 days The 90-day mortality rate in the study population was 12.1% (11/ 91) using KaplaneMeier estimates in Fig. 2, with male mortality at 14.1% (9/64) and female mortality at 7.4% (2/27). The mean age of these patients at death was 73.8 ± 15.4 years [range, 31e87 years]. The average survival period after operation was 36.3 ± 32.6 days [range, 0e87 days]. All patients had severe ASA grade (grade 3e5). A total of 81.8% (9/11) patients underwent AK amputation, and 90.9% (10/11) patients had PAD or diabetes mellitus; 27.3% (3/11) died within 1 week after the operation. In the remainder, who
Table 1 Characteristics of the study population (91 patients with 108 limbs). Variable
Total number of patients/limbs, n (%)
Age (years)a Sex Male Female BMI (kg/m2)a ASA gradea Amputation level AK Total Unilateral Bilateral BK Total Unilateral Bilateral AK after BK Total Unilateral Bilateral Traumatic event Medical comorbidities Hypertension Diabetes mellitus Peripheral arterial disease Liver disease Dialysis Arrhythmia Heart failure Coronary artery disease Smoking Anesthesia General General þ Spinal General þ Block Spinal Block CRP (mg/dL)a WBC (/mL)a HbA1c-NGSP (%)a Missing data
64.9 ± 15.5 64 (70.3)/76 (70.4) 27 (29.7)/32 (29.6) 22.1 ± 4.2 2.84 ± 0.7
57 (62.6)/61 (56.5) 53 (93.0)/53 (86.9) 4 (7.0)/8 (13.1) 42 (46.2)/47 (43.5) 37 (88.1)/37 (78.7) 5 (11.9)/10 (21.3) 8 (8.8)/8 (7.4) 8 (100)/8 (100) e 15 (16.5)/15 (13.9) 48 (52.7)/59 (54.6) 50 (54.9)/60 (55.6) 45 (49.5)/59 (54.6) 9 (9.9)/10 (9.3) 13 (14.3)/22 (20.4) 15 (16.5)/19 (17.6) 16 (17.6)/21 (19.4) 16 (17.6)/19 (17.6) 33 (36.3)/42 (38.9) 62 (68.1)/73 (67.6) 8 (8.8)/11 (10.2) 9 (9.9)/10 (9.3) 7 (7.7)/7 (6.5) 5 (5.5)/7 (6.5) 15.6 ± 11.2 16,747 ± 9311 7.5 ± 2.5 8 (7.7)/8 (7.4)
CRP: C-reactive protein, HbA1c-NGSP: HbA1c-National Glycohemoglobin Standardization Program, WBC: white blood cell count. a The values are given as the mean and standard deviation.
Please cite this article as: Miyamoto S et al., Short-term mortality and risk factors associated with reoperation after emergency amputation of the lower limbs, Journal of Orthopaedic Science, https://doi.org/10.1016/j.jos.2019.05.018
4
S. Miyamoto et al. / Journal of Orthopaedic Science xxx (xxxx) xxx
Table 2 Characteristics of the diabetes mellitus and peripheral arterial disease patients (76 patients with 93 limbs). Variable
Frequency (%) 76 (100)/93 (100)
Number of patients/limbs Age (years)d Sex Male Female BMI (kg/m2)d ASA graded Amputation level AK BK Medical comorbidities Hypertension Liver disease Dialysis Arrhythmia Heart failure Coronary artery disease Smoking CRP (mg/dL)d WBC (/mL)d HbA1c-NGSP (%)d Disruption level of blood flow Aortoiliac Femoropopliteal Infrapopliteal Lack of CTA data
Diabetes mellitus
Peripheral arterial disease
Combined diabetes mellitus and peripheral arterial disease
31 (40.8)/34 (36.6) 62.6 ± 12.0
26 (34.2)/33 (35.5) 72.3 ± 13.7a
19 (25.0)/26 (28.0) 67.8 ± 8.8c
19 (61.3)/22 (64.7) 12 (38.7)/12 (35.3) 23.3 ± 4.3 2.9 ± 0.5
19 (73.1)/24 (72.7) 7 (26.9)/9 (27.3) 19.7 ± 3.5a 2.9 ± 0.6
14 (73.7)/18 (69.2) 5 (26.3)/8 (30.8) 22.1 ± 3.5c 3.0 ± 0.2
15 (48.4)/15 (44.1) 18 (58.1)/19 (55.9)
24 (92.3)/27 (81.8)a 5 (19.2)/6 (18.2)a
10 (52.6)/11 (42.3)c 12 (63.2)/15 (57.7)c
15 (48.4)/18 (52.9) 2 (6.5)/2 (5.9) 1 (3.2)/2 (5.9) 1 (3.2)/1 (2.9) 3 (9.7)/3 (8.8) 2 (6.5)/2 (5.9) 9 (29.0)/11 (32.4) 19.6 ± 12.3 18,212 ± 10,168 9.3 ± 2.8
16 (61.5)/20 (60.6) 3 (11.5)/4 (12.1) 7 (26.9)/11 (33.3)a 10 (38.5)/11 (33.3)a 6 (23.1)/7 (21.2) 8 (30.8)/9 (27.3)a 10 (38.5)/13 (39.4) 13.2 ± 9.9a 15,958 ± 9201 5.5 ± 0.5a
15 (78.9)/19 (73.1)b 2 (10.5)/2 (7.7) 5 (26.3)/9 (34.6)b 4 (21.1)/7 (26.9) 7 (36.8)/11 (42.3)b 6 (31.6)/8 (30.8)b 9 (47.4)/13 (50.0) 15.0 ± 9.4 16,600 ± 9764 8.2 ± 1.7c
e e e e
4 (15.4)/4 (12.1) 16 (61.5)/20 (60.6) 2 (7.7)/2 (6.1) 5 (19.2)/7 (21.2)
1 8 6 6
(5.3)/1 (3.8) (42.1)/11 (42.3) (31.6)/7 (26.9) (31.6)/7 (26.9)
CRP: C-reactive protein, HbA1c-NGSP: HbA1c-National Glycohemoglobin, WBC: white blood cell count, CTA: computed tomography angiography. a Significant difference between diabetes mellitus and peripheral arterial disease (p < 0.05). b Significant difference between diabetes mellitus and combined diabetes mellitus and peripheral arterial disease (p < 0.05). c Significant difference between peripheral arterial disease and combined diabetes mellitus and peripheral arterial disease (p < 0.05). d The values are given as the mean and standard deviation.
Of the 91 patients with 108 limbs, 41.8% (38/91) patients with 40.7% (44/108) limbs experienced postoperative wound complications within 90 days. The overall rate of medical complications within 90 days was 18.7% (17/91) patients, which included pneumonia and urinary tract infection in 6.6% (6/91) patients, sepsis in 4.4% (4/91) patients, heart failure in 1.1% (1/91) patients, renal failure requiring dialysis in 1.1% (1/91) patients, and other complications in 4.4% (4/91) patients. 4. Discussion
Fig. 2. KaplaneMeier survivorship curve with mortality and reoperation as the end point.
survived longer, the average survival period after the operation was 52.1 ± 28.2 days [range, 8e87 days].
3.3. Reoperations and any complication in 90 days The 90-day reoperation rate was 35.2% (32/91) patients with 32.4% (35/108) limbs. Survival analysis with reoperation as the end point is shown in Fig. 2. Univariate Cox regression analyses of 10 parameters for reoperation are summarized in Table 3. All factors with a p value of <0.1 in the univariate analysis were used in multivariate analysis (Table 3). Multivariate analysis revealed that reoperation within 90 days after emergency amputation of the lower limbs was significantly associated with PAD (p ¼ 0.022, HR ¼ 2.75) (Table 4).
To our knowledge, there have been only a few reports about postoperative mortality and reoperation rates after emergency amputation of the lower limbs. Contrary to expectation, the 90-day mortality and reoperation rates after emergency amputation of the lower limbs were high, with a mortality rate of 12.1% and a reoperation rate of 35.2% of patients with 32.4% of limbs. We also found that patients with PAD had a significantly higher risk of reoperation after emergency amputation of the lower limbs. No association was found between laboratory data such as CRP, WBC and HbA1c and reoperation, and furthermore, between diabetes mellitus and reoperation. Also, 41.8% of patients with 40.7% of limbs experienced wound complications and 18.7% of patients with 16.7% of limbs suffered medical complications postoperatively. Since the recent development of endovascular revascularization techniques and bypass surgery, blood circulation of the lower legs was increased in critical limb ischemia and severe trauma. Furthermore, lower leg function and quality of life were greatly improved. However, circulation in the lower limbs could not be salvaged in some elderly patients with multiple complicating comorbidities. In general, the principal objective of amputation is
Please cite this article as: Miyamoto S et al., Short-term mortality and risk factors associated with reoperation after emergency amputation of the lower limbs, Journal of Orthopaedic Science, https://doi.org/10.1016/j.jos.2019.05.018
S. Miyamoto et al. / Journal of Orthopaedic Science xxx (xxxx) xxx
5
Table 3 Univariate analyses of 10 parameters for reoperations (n ¼ 91, p < 0.1). Variable
Number of patients/limbs Age category (years) <60 60e79 80 Sex Male Female BMI (kg/m2) <25 25 ASA grade 1e2 3e5 Amputation level AK BK Traumatic event Medical comorbidities Hypertension Diabetes mellitus Peripheral arterial disease Liver disease Dialysis Arrhythmia Heart failure Coronary artery disease Smoking CRP (mg/dL) <10 10e19 20 Unspecified WBC (/mL) <9000 9000e20000 20,000 HbA1c-NGSP (%) <6.5 6.5e9 10 Unspecified
Frequency (%)
HR (95% CI)
p value
7 (21.9)/8 (22.9) 18 (56.2)/20 (57.1) 7 (21.9)/7 (20.0)
1.0 1.45 (0.64e3.29) 2.06 (0.75e5.68)
e 0.376 0.164
43 (72.9)/53 (72.6) 16 (27.1)/20 (27.4)
21 (65.6)/23 (65.7) 11 (34.4)/12 (34.3)
1.0 1.41 (0.70e2.83)
e 0.336
77 (84.6)/92 (85.2) 14 (15.4)/16 (14.8)
49 (83.1)/61 (83.6) 10 (16.9)/12 (16.4)
28 (87.5)/31 (88.6) 4 (12.5)/4 (11.4)
1.0 0.69 (0.24e1.95)
e 0.48
22 (24.2)/23 (21.3) 69 (75.8)/85 (78.7)
15 (25.4)/15 (20.5) 44 (74.6)/58 (79.5)
7 (21.9)/8 (22.9) 25 (78.1)/27(77.1)
1.0 0.99 (0.45e2.18)
e 0.98
57 (62.6)/61 (56.5) 42 (46.2)/47 (43.5) 15 (16.5)/15 (13.9)
41 (69.5)/43 (58.9) 25 (42.4)/30 (41.1) 8 (13.6)/8 (11.0)
16 (0.5)/18 (51.4) 17 (53.1)/17 (48.6) 7(21.9)/7(20.0)
1.0 1.13 (0.58e2.20) 1.74 (0.76e3.98)
e 0.71 0.192
48 (52.7)/59 (54.6) 50 (54.9)/60 (55.6) 45 (49.5)/59 (54.6) 9 (9.9)/10 (9.3) 13 (14.3)/22 (20.4) 15 (16.5)/19 (17.6) 16 (17.6)/21 (19.4) 16 (17.6)/19 (17.6) 33 (36.3)/42 (38.9)
26 (44.1)/34 (46.6) 34 (57.6)/43 (58.9) 21 (35.6)/32 (43.8) 7 (11.9)/8 (11.0) 6 (10.2)/14(19.2) 6 (10.2)/10 (13.7) 7 (11.9)/11 (15.1) 8 (13.6)/10 (13.7) 23 (39.0)/31 (42.5)
22 (68.8)/25 (71.4) 16 (50.0)/17 (48.6) 24 (75.0)/27 (77.1) 2 (6.3)/2 (5.7) 7 (21.9)/8 (22.9) 9 (28.1)/9 (25.7) 9 (28.1)/10 (28.6) 8 (25.0)/9 (25.7) 10 (31.3)/11(31.4)
2.03 0.62 3.43 0.52 1.12 1.80 2.04 1.84 0.67
(0.97e4.23) (0.32e1.21) (1.56e7.56) (0.12e2.16) (0.51e2.46) (0.84e3.85) (0.98e4.25) (0.86e3.94) (0.33e1.36)
0.059 0.163 0.022 0.365 0.782 0.13 0.058 0.114 0.267
29 (31.9)/40 (37.0) 29 (31.9)/32 (29.6) 32 (35.2)/35 (32.4) 1 (1.1)/1 (0.9)
16 (27.1)/25(34.2) 19 (32.2)/20(27.4) 24 (40.7)/28 (38.4)
14 (43.8)/15 (42.9) 10 (31.3)/12 (34.3) 7 (21.9)/7 (20.0)
1.0 1.03 (0.48e2.20) 0.54 (0.22e1.32)
e 0.941 0.173
12 (13.2)/18 (16.7) 56 (61.5)/62 (57.4) 23 (25.3)/28 (25.9)
8 (13.6)/12(16.4) 34 (57.6)/39(53.4) 17 (28.8)/22(30.1)
4 (12.5)/6 (17.1) 22 (68.7)/23 (65.7) 6 (18.8)/6(17.1)
1.0 1.26 (0.51e3.11) 0.67 (0.22e2.09)
e 0.61 0.494
40 (44.0)/48 (44.4) 30 (33.0)/39 (36.1) 13 (14.2)/13 (12.0) 8 (8.8)/8 (7.4)
26 (44.1)/32 (43.8) 18 (30.5)/26 (35.6) 11 (18.6)/11 (15.1)
14 (43.8)/16 (45.7) 12 (37.5)/13 (37.1) 2 (6.3)/2 (5.7)
1.0 0.86 (0.42e1.80) 0.36 (0.08e1.57)
e 0.696 0.174
Total
No reoperation
Reoperation
91 (100)/108 (100)
59 (64.8)/73 (67.6)
32 (35.2)/35 (32.4)
27 (29.7)/31 (28.7) 46 (50.5)/57 (52.8) 18 (19.8)/20 (18.5)
20 (33.9)/23 (31.5) 28 (47.5)/37 (50.7) 11 (18.6)/13 (17.8)
64 (70.3)/76 (70.4) 27 (29.7)/32 (29.6)
Categorical variables are presented as number (%). Bold p-values are statistically significant. HR: hazard ratio, CI: confidence interval; BMI: body mass index, ASA: American Society of Anesthesiology, AK: above the knee, BK: below the knee, CRP: C-reactive protein, HbA1c-NGSP: HbA1c-National Glycohemoglobin Standardization Program, WBC: white blood cell count.
Table 4 Multivariate analyses of 3 parameters for reoperations (n ¼ 91, p < 0.05). Variable
HR (95%CI)
p value
Hypertension Peripheral arterial disease Heart failure
1.41 (0.64e3.09) 2.75 (1.16e6.53) 1.51 (0.70e3.24)
0.392 0.022 0.289
Bold p-values are statistically significant. HR: hazard ratio, CI: confidence interval.
to improve the pain caused by ischemia, to control the spreading area of infection and gangrene, and to treat septic and hypovolemic shock. A strong association between major amputation and certain comorbidities has been reported. Patients with diabetes mellitus were more likely to need a major amputation, compared to nondiabetic patients with PAD [11,12]. Diabetes mellitus, chronic obstructive pulmonary disease and systemic vasculitis were also associated with a significant risk of lower extremity amputation in severe peripheral arterial occlusive disease [13]. According to several reports, mortality rates after amputation were considerable.
The overall mortality rates after major amputation as described by Pell J et al. were 29% at 6 months, 37% at 1 year, and 49% at 2 years [6]. According to a retrospective study by Klaphake S, the overall mortality rates of patients aged >70 years who were treated for chronic critical limb ischemia by primary or secondary major amputation were 30 days, 14%; 6 months, 38%; 1 year, 44%; 3 years, 66% and 5 years, 85%, respectively [14]. These reports suggested that the short- and long-term mortality rates of these patients were unpredictably high. Although our examination was over a shorter term, our results concur with the other reports. In our study, all patients had ASA grades in the severe range, and almost all patients underwent AK amputation and had PAD or diabetes mellitus as comorbidities. A review of the literature about re-amputation rates after amputation types including toe, ray, midfoot (transmetatarsal amputation, Lisfranc, or Chopart), and major amputations (Syme, transtibial, transfemoral, or hip disarticulation) reported the following rates: 1 year, 26.7%; 3 years, 48.3% and 5 years; 60.7% [15]. Furthermore, the rates of re-amputation on the ipsilateral limb after a major amputation were: 1 year, 4.7%; 3 year, 11.8%; and 5 year, 13.3% [15]. Although our 90-day reoperation rate of 35.2%
Please cite this article as: Miyamoto S et al., Short-term mortality and risk factors associated with reoperation after emergency amputation of the lower limbs, Journal of Orthopaedic Science, https://doi.org/10.1016/j.jos.2019.05.018
6
S. Miyamoto et al. / Journal of Orthopaedic Science xxx (xxxx) xxx
patients with 32.4% limbs was higher than those of reports, these patients might have had a different indication for reoperation, including surgical debridement and re-suture. Our study also found a significant association between PAD and a higher risk of reoperation. Reportedly, the risk factor for an association between reoperation and amputation of the lower limbs is unknown. Hence, our findings may have important implications for the management of amputation of the lower limbs. Our findings should be interpreted in the context of several limitations. First, this was a nonrandomized retrospective study and the sample size was relatively small. Hence, its power for detecting significance among risk factors for reoperation remains low. Second, our study had a short-term follow-up period. The influence of patient characteristics and preoperative comorbidities on long-term mortality should be taken into consideration in future studies. Third, these results were influenced by selection bias; the indication for amputation and the amputation level were decided by individual physicians, and CTA to determine the amputation level was not performed on all patients because some had renal dysfunction in sepsis. Fourth, we did not evaluate medication or fluid management during surgery, or the volume of the surgeons. Fifth, therapeutic options including medication and endovascular revascularization techniques have increased and progressed over time in this study. Furthermore, therapeutic options were decided by individual medical physicians. Hence, the treatment progress leading to amputation varied across the patients. In conclusion, the results of this retrospective case study indicate that the 90-day mortality and reoperation rate after emergency amputation of the lower limbs were high. We also found that patients with PAD were at a significantly higher risk of reoperation. On the other hand, no association was found between laboratory results (CRP, WBC and HbA1c) or diabetes mellitus and reoperation. This study contributes to the literature, as these findings will have important implications for treatment of lower leg amputation and prevention of reoperation. Furthermore, the study findings can provide a useful management strategy for not only orthopedic trauma surgeons, but also general surgeons. Conflicts of interest The authors declare that there are no conflicts of interest regarding the publication of this paper. Acknowledgements We are grateful to Dr Makoto Otsuka, Dr Fumio Hasue, Dr Takayuki Fujiyoshi, Dr Koushirou Kamiya, Dr Hitoshi Kiuchi, Dr Tadashi Tanaka, Dr Junichi Nakamura, Dr Sumihisa Orita, and Professor Seiji Ohtori, head of our department, for their valuable suggestions and support during the course of this study. All authors
have read and approved the submitted version. Study design: Shuichi Miyamoto, Makoto Otsuka, Tadashi Tanaka, Junichi Nakamura, Sumihisa Orita, Seiji Ohtori. Drafting the protocol: Shuichi Miyamoto, Makoto Otsuka, Tadashi Tanaka, Sumihisa Orita, Seiji Ohtori. Advice on the statistical analysis: Shuichi Miyamoto, Sumihisa Orita. Patient recruitment and data collection: Shuichi Miyamoto, Fumio Hasue, Takayuki Fujiyoshi, Koushirou Kamiya, Hitoshi Kiuchi. First draft of the manuscript: Shuichi Miyamoto. We did not receive any funding or financial support that may be perceived to have biased the study.
References [1] Jude EB, Oyibo SO, Chalmers N, Boulton AJ. Peripheral arterial disease in diabetic and nondiabetic patients: a comparison of severity and outcome. Diabetes Care 2001;24(8):1433e7. [2] Brownrigg JR, Apelqvist J, Bakker K, Schaper NC, Hinchliffe RJ. Evidence-based management of PAD & the diabetic foot. Eur J Vasc Endovasc Surg 2013;45(6): 673e81. [3] Norgren L, Hiatt WR, Dormandy JA, Nehler MR, Harris KA, Fowkes FG. TASC II working group. Inter-Society consensus for the management of peripheral arterial disease (TASC II). J Vasc Surg 2007;45(Suppl S):S5e67. [4] Reinecke H, Unrath M, Freisinger E, Bunzemeier H, Meyborg M, Lüders F, Gebauer K, Roeder N, Berger K, Malyar NM. Peripheral arterial disease and critical limb ischaemia: still poor outcomes and lack of guideline adherence. Eur Heart J 2015;36(15):932e8. [5] Dormandy JA. Natural history of patients with peripheral occlusive arterial disease. Ann Chir Gynaecol 1992;81(2):86e8. [6] Pell J, Stonebridge P. Association between age and survival following major amputation. The Scottish vascular audit group. Eur J Vasc Endovasc Surg 1999;17(2):166e9. [7] Johansen K, Daines M, Howey T, Helfet D, Hansen Jr ST. Objective criteria accurately predict amputation following lower extremity trauma. J Trauma 1990;30(5):568e72. [8] Terry Canale S, James H. In: Beaty Campbell's operative orthopaedics. 12th ed. Philadelphia: Elsevier Mosby; 2013. p. 637e50 [Amputations of the lower extremity]. [9] Saklad M. Grading of patients for surgical procedures. Anesthesiology 1941;2(3):281e4. [10] Dripps RD, Lamont A, Eckenhoff JE. The role of anesthesia in surgical mortality. JAMA 1961;178(3):261e6. [11] Carmona GA, Hoffmeyer P, Herrmann FR, Vaucher J, Tschopp O, Lacraz A, Vischer UM. Major lower limb amputations in the elderly observed over ten years: the role of diabetes and peripheral arterial disease. Diabetes Metab 2005;31(5):449e54. hn AB, Atroshi I. [12] Johannesson A, Larsson GU, Ramstrand N, Turkiewicz A, Wire Incidence of lower-limb amputation in the diabetic and nondiabetic general population: a 10-year population-based cohort study of initial unilateral and contralateral amputations and reamputations. Diabetes Care 2009;32(2): 275e80. [13] Lu MC, Hsu H, Lin CH, Koo M. Systemic vasculitis is associated with a higher risk of lower extremity amputation in patients with severe peripheral arterial occlusive disease: a secondary analysis of a nationwide, population-based health claims database. Rheumatol Int 2017;37(11):1847e52. [14] Klaphake S, de Leur K, Mulder PG, Ho GH, de Groot HG, Veen EJ, Verhagen HJ, van der Laan L. Mortality after major amputation in elderly patients with critical limb ischemia. Clin Interv Aging 2017;12:1985e92. [15] Izumi Y, Satterfield K, Lee S, Harkless LB. Risk of reamputation in diabetic patients stratified by limb and level of amputation: a 10-year observation. Diabetes Care 2006;29(3):566e70.
Please cite this article as: Miyamoto S et al., Short-term mortality and risk factors associated with reoperation after emergency amputation of the lower limbs, Journal of Orthopaedic Science, https://doi.org/10.1016/j.jos.2019.05.018