- Email: [email protected]
Venous thromboembolism after elective shoulder surgery: a prospective cohort study of 175 patients
J Shoulder Elbow Surg (2014) 23, 605-612
www.elsevier.com/locate/ymse
JAPAN SHOULDER SOCIETY ARTICLES
Venous thromboembolism after elective shoulder surgery: a prospective cohort study of 175 patients Hiroyuki Takahashi, MDa, Nobuyuki Yamamoto, MD, PhDa, Hideaki Nagamoto, MDa, Hirotaka Sano, MD, PhDa, Minoru Tanaka, MDb, Eiji Itoi, MD, PhDa,* a b
Department of Orthopaedic Surgery, Tohoku University School of Medicine, Sendai, Japan Department of Orthopaedic Surgery, Tohoku Rosai Hospital, Sendai, Japan Background: The purpose of this study was to investigate the incidence of venous thromboembolism (VTE) after elective arthroscopic shoulder surgery. Materials and methods: One hundred seventy-five consecutive patients who underwent arthroscopic shoulder surgery were enrolled (mean age, 61 years). Patients who had VTE preoperatively and underwent trauma surgery or arthroplasty were excluded. All the patients used foot pumps or elastic stockings after surgery for deep venous thrombosis (DVT) prophylaxis. DVT in the 4 limbs was assessed by ultrasound before and after surgery. Pulmonary embolism was diagnosed by computed tomography pulmonary angiography. Risk factors related to DVT were assessed. Results: The overall incidence of DVT was 10 of 175 patients (5.7%). Most of the DVT cases were detected at 1 to 2 days after surgery. All patients were asymptomatic. There were no patients who had symptomatic pulmonary embolism. However, an asymptomatic pulmonary embolus developed in 1 patient during the 3-month follow-up period. There were no significant differences between the DVT and nonDVT groups regarding the risk factors. Discussion and conclusion: Our data have shown that symptomatic VTE is rare after elective arthroscopic shoulder surgery. However, asymptomatic VTE may occur even with DVT preventive measures. Because most of the DVTs were found in the calf veins, we recommend that surgeons pay attention to the possibility of DVT in the lower extremities even after arthroscopic shoulder surgery. The incidence of asymptomatic VTE after elective arthroscopic shoulder surgery was 5.7%. All patients were asymptomatic, and most of the DVTs occurred 1 or 2 days after surgery. Level of evidence: Level I, Prospective Cohort, Prognosis Study. Ó 2014 Journal of Shoulder and Elbow Surgery Board of Trustees. Keywords: Venous thromboembolism; deep venous thrombosis; pulmonary embolism; shoulder surgery; arthroscopy; asymptomatic; ultrasonography
This study was approved by the institutional review board of Tohoku University Graduate School of Medicine (study No. 2011-250, 2012-1-96). *Reprint requests: Eiji Itoi, MD, PhD, Department of Orthopaedic Surgery, Tohoku University School of Medicine, 1-1, Seiryo-machi, Aoba-ku, Sendai, Japan 980-8574. E-mail address: [email protected] (E. Itoi).
The incidence of calf deep venous thrombosis (DVT) after total knee or hip arthroplasty has been reported to be 20% to 40%.17 Guidelines for the diagnosis, treatment, and prevention of venous thromboembolism (VTE) after total knee or hip arthroplasty and open surgery for
1058-2746/$ - see front matter Ó 2014 Journal of Shoulder and Elbow Surgery Board of Trustees. http://dx.doi.org/10.1016/j.jse.2014.01.054
606 proximal femoral fracture have been published. The guidelines include DVT standard prophylaxis such as intermittent pneumatic compression devices, elastic stockings, and anticoagulation therapy for perioperative management.11,15,17,18 According to the DVT guidelines published by the Japanese Circulation Society in 2009, patients undergoing surgery of the upper extremity are not at risk for postoperative DVT if there are no risk factors related to DVT.17 Preventive measuresdwith the exception of early ambulation and active calf stretch exercisesdare not recommended.11,18 In 1990, Burkhart5 reported the first case of symptomatic DVT after shoulder arthroscopic surgery. Meanwhile, the number of reports about symptomatic VTE after shoulder surgery has been gradually increasing.1,3,4,7,8,10,12-14,16,19-24,27 In 2009, Willis et al27 reported that the prevalence of DVT after shoulder arthroplasty was 13% in 100 patients. Although there has been increased awareness of the existence of DVT after open shoulder surgery, it still has not been a major concern for shoulder surgeons after arthroscopic shoulder surgery. Some retrospective studies have shown that the incidence of symptomatic VTE events after shoulder surgery was extremely low, ranging from 0.01% to 0.68%.8,10,16,20,23 However, there have been only 2 prospective studies investigating the incidence of VTE after elective shoulder surgery.26,27 The purposes of this study were (1) to prospectively investigate the incidence of VTE after elective arthroscopic shoulder surgery and (2) to clarify the risk factors related to VTE events. Our hypothesis was that symptomatic VTE events after elective arthroscopic shoulder surgery are rare whereas asymptomatic VTE might occur even with DVT prophylaxis.
Materials and methods This is a prospective cohort study of VTE that occurred after elective arthroscopic shoulder surgery.
Patients A prospective cohort study was designed to document the incidence of VTE after shoulder surgery in consecutive patients. Between June 2011 and May 2013, a total of 353 consecutive shoulder surgeries were performed at our hospital, as well as a related hospital (one of our affiliated hospitals). Of the patients, 175 who met the following inclusion criteria were enrolled: (1) patients underwent elective arthroscopic shoulder surgery and (2) patients provided informed consent. The exclusion criteria were (1) patients who had a history of symptomatic VTE preoperatively, (2) trauma patients including elective fracture surgery cases, (3) patients who underwent arthroplasty, and (4) patients who were unable to undergo ultrasound examination before and after surgery. Arthroscopic surgery was performed in 175 shoulders in 175 patients. There were 145 arthroscopic rotator cuff surgeries, 17 arthroscopic Bankart repairs, and 13 other
H. Takahashi et al. procedures (Table I). There were 125 men and 50 women, and the mean age at the time of surgery was 61 13 years (range, 1880 years). All procedures were performed with patients under general anesthesia, in the beach-chair position. The procedure was not conducted in the lateral decubitus position in any patient.
DVT preventive measures During and after surgery, DVT prophylaxis such as elastic stockings, pneumatic compression pumps, and early ambulation was used in all patients. Intermittent pneumatic compression devices were used as DVT prophylaxis for the lower extremities during surgery in 173 cases (99%). An elastic stocking for the lower extremities was used in 2 of 175 cases (1%) during surgery. None of the patients in this study used low–molecular weight heparin or aspirin. The pneumatic compression pumps were used from the beginning of surgery for 24 hours. Patients who were not able to begin early ambulation used pneumatic compression pumps until they started ambulation. On the day after surgery, 171 patients (98%) were able to get out of bed. After surgery, intermittent pneumatic compression devices were used in 162 patients (93%), elastic stockings in 2 (1%), and the combination thereof in 11 (6%).
Questionnaire and ultrasound examination A questionnaire was completed before surgery. The potential risk factors associated with VTE, which have been previously reported, were assessed.2,17,27 As preoperative factors, the patient’s background (age, sex, race, body mass index [BMI], and smoking habit), the presence of malignant disease, and comorbidities diagnosed previously were documented (Table II). As perioperative and postoperative factors, operative time, complications, length of hospitalization, and any associated VTE complications, such as symptomatic or fatal pulmonary embolism (PE), were recorded. A 4-limb surveillance duplex ultrasound examination was performed to investigate the incidence of DVT. All examinations were performed and interpreted by a single orthopaedic surgeon specializing in musculoskeletal ultrasonography. In addition, he had 3 months of training in venous ultrasonography to detect DVT in the 4 limbs. To show the accuracy of detecting DVT by ultrasound in this study, the incidence of lower-extremity DVT in patients who underwent total knee arthroplasty during the same period was also investigated by conducting the same ultrasound procedure. Thirty-nine patients who had total knee arthroplasty at our hospital were enrolled, comprising 7 male and 32 female patients with a mean age of 72 years. Their diagnoses were osteoarthritis in 31 cases, rheumatoid arthritis in 7 cases, and psoriatic arthritis in 1 case. The ultrasound examination was performed between 3 days and 1 month after total knee arthroplasty. The ultrasound devices used in this study were the Hi-Vision Preirus system (Hitachi-Aloka Medical, Tokyo, Japan) and MTurbo system (SonoSite, Bothell, WA, USA) with linear probes ranging from 6 to 13 MHz and convex probes ranging from 2 to 5 MHz. The posture was the supine position with the arm in slight abduction. The examination sites were from the elbow joint level to the subclavian vein in the upper extremity and from the posterior calf to the inguinal area in the lower extremity. Duplex
Venous thromboembolism after shoulder surgery Table I
607
Demographic data of all patients
Demographic data No. of patients Age (range) (y) Male-female ratio Diagnosis (n) RCT RAD Others
Overall
Arthroscopic rotator cuff surgery ARCR
A-Patch
ABR
Others
175 61 13 (18-80) 2.5:1
134 61 14 (18-87) 2.5:1
11 69 7 (55-78) 2.7:1
17 38 16 (18-66) 3.3:1
13 60 14 (34-76) 1.6:1
149 (85%) 17 (10%) 9 (5%)
130 (97%) 0 (0%) 4 (3%)
11 (100%) 0 (0%) 0 (0%)
0 (0%) 17 (100%) 0 (0%)
8 (62%) 0 (0%) 5 (38%)
A-Patch, Arthroscopic patch graft procedure; ABR, arthroscopic Bankart repair; ARCR, arthroscopic rotator cuff repair; RAD, recurrent anterior dislocation; RCT, rotator cuff tear.
Table II
Potential VTE risk factors and perioperative data Data
VTE risk factors (n) Age 60 y BMI 30 kg/m2 Comorbidity Diabetes mellitus Hypertension Cardiovascular COPD CKD Chronic liver failure Active cancer Inflammatory disease Smoking habit History/other factors Family history of VTE Cancer Cerebral stroke VTE Major surgery Orthopaedic surgery Trauma (within 6 mo) Additional surgery Perioperative data Operative time (min) Total anesthesia time (min) Length of hospital stay (d)
111 (63%) 7 (4%) 34 (19%) 62 (35%) 14 (8%) 10 (6%) 2 (1%) 3 (2%) 2 (1%) (prostate) 3 (2%) (RA in 2 and Behc¸et disease in 1) 81 (46%) 0 2 8 0 28 33 2 8
(0%) (1%) (5%) (0%) (16%) (19%) (1%) (5%)
135 (95% CI, 126-145) 214 (95% CI, 204-223) 33 (95% CI, 30-35)
CI, Confidence interval; CKD, chronic kidney disease; COPD, chronic obstructive pulmonary disease; RA, rheumatoid arthritis.
ultrasound has been reported to be an accurate and effective screening modality.29 DVT was diagnosed by ultrasound when both of the following findings were present: (1) intravenous hyperechoic lesion or venous incompressibility on B-mode imaging and (2) venous flow defect or abnormality proven in colorflow Doppler mode. When DVT was diagnosed by ultrasound, the serum D-dimer level was also tested to establish the differential diagnosis. Both DVT findings by ultrasound and positive D-dimer
levels indicated a definitive diagnosis of VTE.29 In cases with DVT in which the clinical symptoms of PE were suspected, computed tomography pulmonary angiography (CTPA) was conducted to confirm PE.13 The presence and location of DVT, as well as clinical symptoms related to VTE, were recorded. The mean follow-up period was 10 months after surgery, and DVT after hospital discharge was assessed by the operating surgeon.
Timing of ultrasonographic diagnosis Most thromboembolic events after lower-extremity surgery have been reported to occur in the first postoperative week.28 Because it has been reported that DVT and related thromboembolic complications occur until 12 weeks after knee or hip arthroplasty, it was necessary for us to carefully observe whether VTE events occurred until 3 months after surgery.11,18 Thus, duplex ultrasound was performed (1) before surgery, (2) 1 or 2 days after surgery (acute phase), and (3) 3 weeks to 3 months after surgery (subacute phase). All 175 patients underwent preoperative and postoperative acute-phase examinations, but only 85 patients underwent subacute-phase examinations because some patients or the examiner had time restrictions regarding the ultrasonographic examinations in the outpatient clinic.
Statistical analysis We used c2 analysis or the Fisher exact test to investigate the difference in the incidence of DVT with regard to sex, surgical procedures, smoking habit, and comorbidity. We performed the Mann-Whitney U test to assess the difference in the incidence of DVT with regard to age, BMI, operative time, and other factors. No corrections were made for multiple comparisons. All statistical analyses were performed with JMP Pro 9 software (SAS Institute, Cary, NC, USA). When this study was planned, there was only 1 report available in the literature on the same topic.27 Therefore, it was difficult for us to calculate the estimated sample size. The sample size was determined based on the number of shoulder surgeries that were performed at our hospitals (about 200 cases per year). The power analysis was calculated after this study to check whether the hypothesis was correct. We used G*power 3 software to calculate the statistical power.9
608
H. Takahashi et al.
Figure 1 (A) Ultrasonographic longitudinal image of upper-extremity DVT in a 69-year-old woman, 4 weeks after an arthroscopic patch graft procedure. A fresh free-floating DVT was found in the affected subclavian vein (arrows) on the longitudinal image by ultrasonographic B-mode imaging. (B) Ultrasonographic axial image of upper-extremity DVT. A defect of venous flow was seen in the affected subclavian vein (arrows) on the axial image in ultrasonographic color-flow Doppler mode. SCA, Subclavian artery. (C) Computed tomography pulmonary angiogram. Asymptomatic PE was detected in the right pulmonary artery (arrow).
Results Incidence of DVT The overall incidence of DVT was 10 of 175 patients (5.7%). However, all patients had no symptoms related to VTE. DVT was found in the upper (subclavian vein) and lower (calf vein) extremities in 1 patient and in the lower extremities in 9 patients. DVT was detected in 10 cases at 1 to 2 days after surgery and in 1 case at 3 weeks to 3 months after surgery. The DVT observed in the subclavian vein was a free-floating thrombus (Figs 1, A and B). All DVTs that occurred in the lower extremities were the distal type of DVT. Nine cases had DVTs in the soleus veins, and 1 case had DVT in the gastrocnemius vein. In this study, proximal DVT including popliteal and femoral vein thrombosis was not found. Bilateral calf vein thromboses were detected in 4 cases (Table III). Two patients underwent CTPA. DVTs were found in the bilateral soleus veins on the day after surgery in 1 case. This patient showed a reduction in percutaneous oxygen saturation; this prompted the cardiologist to perform the CTPA examination, without any positive findings.
follow-up. This patient had a DVT in the upper and lower extremities. This asymptomatic and fresh DVT occurred in the subclavian vein of the operative upper extremity at around 4 weeks after an arthroscopic patch graft procedure for a massive rotator cuff tear. Then, an asymptomatic PE was detected by CTPA (Fig 1, C). This patient needed to receive anticoagulant therapy with warfarin for 6 months.
Incidence of VTE after total knee arthroplasty The incidence of lower-extremity DVT in patients who underwent total knee arthroplasty during the same period was 13 of 39 cases (33%). All DVTs were found in the calf veins and were asymptomatic. No symptomatic PE was found.
DVT risk factors The patients whose DVT had been proved by ultrasound (DVT group) and the patients who had no DVT (non-DVT group) were compared. There were no significant differences between the DVT and non-DVT groups regarding the risk factors, such as age, sex, BMI, operative time, smoking habit, and comorbidity (Table IV).
Incidence of PE
Ex post analysis of statistical power
There were no patients who had symptomatic PE in this study. However, an asymptomatic pulmonary embolus developed, as evidenced on CTPA, in 1 patient at 3 months’
Because this study could not predict the risk factors from the incidence of VTE after shoulder surgery, the power analyses after the fact were calculated on the risk factors
Venous thromboembolism after shoulder surgery
609
Table III
Clinical features of VTE patients after shoulder surgery
Age (y)
Sex
Diagnosis
Operative procedure
DVT location
Findings and timing of US (initially postoperatively, at final follow-up)
PE
VTE risk factors
79
M
RCT
ARCR
Bilateral soleus
No
HT
75
M
RCT
ARCR
Bilateral soleus
No
HT
63
M
RCT
ARCR
No
d
58 66
M M
RCT RAD
ARCR ABR
Contralateral gastrocnemius Ipsilateral soleus Contralateral soleus
DVT positive at 1 d, DVT negative at 11 d DVT positive at 2 d, DVT positive at 1 wk DVT positive at 1 d, no follow-up
No No
HT HT
64 70
F M
RCT RCT
ARCR ARCR
Ipsilateral soleus Bilateral soleus
No No
60
M
RCT
ARCR
Ipsilateral soleus
No
HT, DM Prostate cancer d
60
M
RCT
ARCR
Contralateral soleus
No
d
69
F
RCT
A-Patch
Bilateral soleus Affected subclavian
No Yes
HT
DVT positive at 2 d, no follow-up DVT positive at 2 d, DVT negative at 9 d DVT positive at 2 d, no follow-up DVT positive at 1 d, DVT negative at 3 wk DVT positive at 1 d, DVT negative at 3 d DVT positive at 1 d, DVT negative at 3 mo DVT positive at 2 d DVT positive at 4 wk
A-Patch, Arthroscopic patch graft procedure; ABR, arthroscopic Bankart repair; ARCR, arthroscopic rotator cuff repair; DM, diabetes mellitus, F, female; HT, hypertension; M, male; RAD, recurrent anterior dislocation; RCT, rotator cuff tear; US, ultrasonography.
retrospectively. Calculating statistical power for VTE risk factors, we found that the power was in the range between 0.098 and 0.355, which was not sufficient to make any conclusion on the risk factors.
Discussion Although it has been reported that symptomatic VTE after shoulder surgery is a rare complication, there have been no reports describing asymptomatic VTE. Recently, some reports describing VTE after shoulder surgery have been published.1,3,4,7,8,10,12-14,16,19-24,26,27 In addition, it has been pointed out that there is a risk of asymptomatic DVT even after arthroscopic surgery, although the risk of DVT has been reported to be at low.25 However, the risk of VTE after shoulder surgery is very low because there are no highevidence reports about VTE after shoulder surgery according to the guidelines.17 Willis et al27 reported that the prevalence of asymptomatic DVT after shoulder surgery was 13% and the rate of symptomatic PE was 3%. However, all surgeries were arthroplasty cases. In our study, the incidence of asymptomatic DVT after surgery was 5.7% and the proportion of symptomatic PE was 0%. However, asymptomatic PE was detected by CTPA in 1 case. All surgeries in our study were arthroscopic surgeries. The overall incidence of DVT in this study was 5.7% in 175 patients. This difference may come from the different inclusion criteria for the different surgical procedures. Willis et al included patients with a history of VTE events that have been reported as yielding the highest risk of DVT in
the guidelines. All the procedures in their series consisted of shoulder arthroplasties. Our data showed that DVT occurred in all cases at 1 or 2 days after surgery. In contrast to the report of Willis et al, arthroscopic surgeries were performed in all of our cases and a history of VTE was an exclusion criterion in our study. Because of this, the incidence of VTE in this study appears lower than that in the previous report. Only 1 patient underwent anticoagulant treatment with heparin for a few days, because he showed a transient decrease in percutaneous oxygen saturation on monitoring, but this was discontinued after a CTPA showed no PE. Only 1 patient underwent anticoagulant treatment with heparin for a few days until CTPA showed no PE because he showed a transient decrease in percutaneous oxygen saturation on monitoring. In another case, no treatment was performed. Of the 10 cases with DVT, 7 (70%) were followed up by repeat ultrasonography. In 5 of these 7 cases (71%), the DVT findings changed during the follow-up period. Among the cases in which DVT was detected by ultrasound 1 to 2 days after surgery, DVT disappeared between 3 days and 3 months after surgery in 5 of 10 cases (50%) in this study. In the literature, there have been no reports investigating the incidence of DVT longitudinally except for the report of Willis et al. Regarding lower-extremity surgery, Cullison et al6 reported that DVT was detected in only 1 of 67 cases and that asymptomatic DVT was seen in the femoral vein 3 days after surgery but disappeared 10 days after surgery. Willis et al27 conducted ultrasonographic evaluation of 100 patients at 2 days after surgery. Half of these patients (50 patients) were examined at 12 weeks after surgery.
610 Table IV
H. Takahashi et al. Comparison between DVT group and non-DVT group
VTE risk factors
DVT group
Non-DVT group
P value
Power
Mean age (y) Male-female ratio Mean BMI (kg/m2) Mean operative time (min) % with smoking habit % with DVT risk–related comorbidity
66 7 4:1 24.3 3.8 139 (95% CI, 107-171) 70 80
60 14 2.4:1 24.3 3.3 135 (95% CI, 125-145) 45 63
.258 .726 >.99 .507 .191 .334
0.212 0.101 d 0.098 0.355 0.185
CI, Confidence interval.
However, Willis et al did not check for DVTs preoperatively. In their study, because preoperative ultrasound examination was not performed, they cannot completely deny the possibility of pre-existing DVTs. Accordingly, they used the term ‘‘prevalence’’ in their report. In our study, on the other hand, ultrasound examination was performed 3 times: before surgery, 1 to 2 days after surgery (acute phase), and 3 weeks to 3 months after surgery (subacute phase). Thus, we were able to determine the incidence of DVT. These differences may also be related to the difference in the incidences between our study and that of Willis et al. Widmer et al26 reported that the prevalence rates of both DVT and PE after shoulder surgery for proximal humeral fracture were 0%. Their DVT preventive measures were aspirin administration and use of pneumatic compression pumps. The timing of their ultrasonography was, on average, 14 days after surgery, which was earlier than that in our study and in the study by Willis et al.27 The possibility of not being able to detect early DVT after surgery could be a major concern in the study by Widmer et al. They excluded patients who had active cancer and a hypercoagulable state, which have been known to yield a high risk of postoperative VTE. Their finding of a 0% prevalence rate might be caused by the late timing of ultrasonography, the inclusion criteria, and the small sample size of 50 patients. It remains unclear when DVT after surgery occurs in the subacute phase. Therefore, our detection of DVT may have been benefited by defining the subacute phase with a wide range, from 3 weeks to 3 months after surgery in this study. Jameson et al16 reported a large-scale analysis comprising 80,227 patients from a national database. They showed that the overall prevalence of symptomatic VTE events after open shoulder surgery was less than 0.7% and the overall prevalence after arthroscopic procedure was 0.01%. Kuremsky et al19 investigated 1,908 cases of shoulder arthroscopic surgery retrospectively. They found that 6 patients (0.31%) were diagnosed with symptomatic VTE by ultrasound and computed tomography. There have been several reports describing the rarity of symptomatic VTE events after shoulder arthroscopic procedure, and the incidences ranged from 0.01% to 0.38%.3,4,10,14,16,20,23 The incidence of VTE after arthroscopic surgery in our study was 10 of 175 cases (5.7%). This value is higher than the
values in the previous reports, although all VTEs were asymptomatic in our study. To validate the accuracy of ultrasonographic diagnosis for DVT in this study, the incidence of lower-extremity DVT in patients who underwent total knee arthroplasty during the same period was investigated. DVTs were found in 13 of 39 patients (33%), and all were asymptomatic. It has been reported that the incidence of DVT after total knee arthroplasty was 36% on average (range, 25%-59%).28 Because the incidence of DVT in this study was close to our incidence, we assumed that the accuracy of diagnosis of DVT in this study was validated to be at an acceptable level. The surgical position and the anatomic structure of the calf veins have been thought to be a possible cause of DVT formation after shoulder surgery. In our study, soleus vein thromboses were found in 9 of 10 cases. The anatomic and physiological features of the soleus veins are that they are venous sinuses that have a plenty of bypass to the surrounding veins such as gastrocnemius and posterior tibial veins, being predisposed to excessive venous congestion in the lower extremity. The calf vein has been reported as the most common site for DVT observed after lower-extremity surgery.28 With patients in the beach-chair position, the hip and knee joints were always flexed during surgery. This may induce DVT formation in the lower extremities during surgery. Because no surgeries were performed with patients in the lateral decubitus position in our series, the difference between these positions remains unclear. Further studies are necessary to show the difference between these positions. All 10 cases of DVT in our study were detected at 1 to 2 days after surgery. This is similar to the report of Willis et al.27 From this result, we are able to conclude that we need to carefully observe whether VTE events occur soon after shoulder surgery. Considering the fact that most DVTs occur in the calf veins, surgeons should be focused on preventive care for calf DVTs after shoulder surgery regardless of whether VTE-related symptoms are present. There are several limitations in this study. First, the lateral decubitus position was not included. Thus, the incidences of VTE could not be compared between the beach-chair and lateral decubitus positions. Second, DVT prophylaxis during and after surgery was not controlled: elastic stockings, pneumatic compression pumps, and early ambulation were used as
Venous thromboembolism after shoulder surgery DVT prophylaxis. However, intermittent pneumatic compression devices were used in 99% of cases during surgery and in 99% of cases after surgery. Thus, we believe that the effect of variable prophylaxis was small. Third, ultrasonography was performed in only 85 patients at 3 weeks to 3 months after surgery, which was approximately half of all the patients. When we divided the patients into 2 groupsdthat is, the single ultrasonography group, comprising the 90 patients who underwent ultrasound at only 1 to 2 days after surgery, and the followup ultrasonography group, comprising the 85 patients who were able to undergo both acute- and subacute-phase examinations postoperativelydthere were significant differences between the 2 groups in age, diagnosis, operative time, and length of hospital stay. Therefore, it was a limitation of this study not to be able to perform ultrasonographic examination at the same time points. Fourth, ultrasound examination for patients who had total knee arthroplasty was performed at 3 days after surgery or later. Because closed drainage with a compressive bandage was routinely used until 2 days after surgery, ultrasound examination could not be performed during the early postoperative period. Lastly, the ex post analysis of the statistical power for DVT risk factors showed that all power values were lower than 0.4, which was considered insufficient to detect any risk factor in this study. Further prospective studies including greater numbers of patients are needed to determine the risk factors for DVT after elective shoulder surgery.
Conclusions The incidence of VTE after elective arthroscopic shoulder surgery was 5.7%. All patients who had DVTs were asymptomatic, and most of the DVTs occurred 1 or 2 days after surgery.
Acknowledgment We thank Professor Takuhiro Yamaguchi, United Centers for Advanced Research and Translational Medicine, Tohoku University Graduate School of Medicine, for his assistance with the statistical analyses. We appreciate the support of Dr Katsumi Sato, Director, and Dr Ko Himori, Tohoku Rosai Hospital. Our grateful thanks go to the staff of Tohoku University Hospital and Tohoku Rosai Hospital for their assistance in collecting and processing data for the study.
Disclaimer The authors, their immediate families, and any research foundations with which they are affiliated have not received any financial payments or other benefits from
611 any commercial entity related to the subject of this article.
References 1. Amarasekera SS, van Dalen J, Thompson TJ, Osman M. Pulmonary embolism after acromioplasty and rotator cuff repair. J Shoulder Elbow Surg 2008;17:e13-4. http://dx.doi.org/10.1016/j.jse.2008.02.001 2. Anakwe RE, Middleton SD, Beresford-Cleary N, McEachan JE, Talwalkar SC. Preventing venous thromboembolism in elective upper limb surgery. J Shoulder Elbow Surg 2013;22:432-8. http://dx.doi.org/ 10.1016/j.jse.2012.10.033 3. Bongiovanni SL, Ranalletta M, Guala A, Maignon GD. Case reports: heritable thrombophilia associated with deep venous thrombosis after shoulder arthroscopy. Clin Orthop Relat Res 2009;467:2196-9. http:// dx.doi.org/10.1007/s11999-009-0895-6 4. Brislin KJ, Field LD, Savoie FH. Complications after arthroscopic rotator cuff repair. Arthroscopy 2007;23:124-8. http://dx.doi.org/10. 1016/j.arthro.2006.09.001 5. Burkhart SS. Deep venous thrombosis after shoulder arthroscopy. Arthroscopy 1990;6:61-3. 6. Cullison TR, Muldoon MP, Gorman JD, Goff WB. The incidence of deep venous thrombosis in anterior cruciate ligament reconstruction. Arthroscopy 1996;12:657-9. 7. Dattani R, Smith CD, Patel VR. The venous thromboembolic complications of shoulder and elbow surgery: a systematic review. Bone Joint J 2013;95-B:70-4. http://dx.doi.org/10.1302/0301-620X.95B1. 29854 8. Farng E, Zingmond D, Krenek L, Soohoo NF. Factors predicting complication rates after primary shoulder arthroplasty. J Shoulder Elbow Surg 2011;20:557-63. http://dx.doi.org/10.1016/j.jse.2010.11. 005 9. Faul F, Erdfelder E, Lang AG, Buchner A. G*Power 3: a flexible statistical power analysis program for the social, behavioral, and biomedical sciences. Behav Res Methods 2007;39:175-91. http://dx. doi.org/10.3758/BF03193146 10. Garofalo R, Notarnicola A, Moretti L, Moretti B, Marini S, Castagna A. Deep vein thromboembolism after arthroscopy of the shoulder: two case reports and a review of the literature. BMC Musculoskelet Disord 2010;11:65. http://dx.doi.org/10.1186/1471-247411-65 11. Geerts WH, Bergqvist D, Pineo GF, Heit JA, Samama CM, Lassen MR, et al. Prevention of venous thromboembolism: American College of Chest Physicians evidence-based clinical practice guidelines (8th edition). Chest 2008;133:381S-453S. http://dx.doi.org/10. 1378/chest.08-0656 12. Greene JW, Deshmukh AJ, Cushner FD. Thromboembolic complications in arthroscopic surgery. Sports Med Arthrosc 2013;21:69-74. http://dx.doi.org/10.1097/JSA.0b013e31828a7e76 13. Hoxie SC, Sperling JW, Cofield RH. Pulmonary embolism after operative treatment of proximal humeral fractures. J Shoulder Elbow Surg 2007;16:782-3. http://dx.doi.org/10.1016/j.jse.2006.12.004 14. Hoxie SC, Sperling JW, Cofield RH. Pulmonary embolism following rotator cuff repair. Int J Shoulder Surg 2008;2:49-51. http://dx.doi.org/ 10.4103/0973-6042.42576 15. Izquierdo R, Voloshin I, Edwards S, Freehill MQ, Stanwood W, Wiater JM, et al. American academy of orthopaedic surgeons clinical practice guideline on: the treatment of glenohumeral joint osteoarthritis. J Bone Joint Surg Am 2011;93:203-5. 16. Jameson SS, James P, Howcroft DW, Serrano-Pedraza I, Rangan A, Reed MR, et al. Venous thromboembolic events are rare after shoulder surgery: analysis of a national database. J Shoulder Elbow Surg 2011; 20:764-70. http://dx.doi.org/10.1016/j.jse.2010.11.034
612 17. JCS Joint Working Group. Guidelines for the diagnosis, treatment and prevention of pulmonary thromboembolism and deep vein thrombosis (JCS 2009). Circ J 2011;75:1258-81. http://dx.doi.org/10.1253/circj. CJ-88-0010 18. Johanson NA, Lachiewicz PF, Lieberman JR, Lotke PA, Parvizi J, Pellegrini V, et al. American academy of orthopaedic surgeons clinical practice guideline on. Prevention of symptomatic pulmonary embolism in patients undergoing total hip or knee arthroplasty. J Bone Joint Surg Am 2009;91:1756-7. http://dx.doi.org/10.2106/JBJS.I.00511 19. Kuremsky MA, Cain EL, Fleischli JE. Thromboembolic phenomena after arthroscopic shoulder surgery. Arthroscopy 2011;27:1614-9. http://dx.doi.org/10.1016/j.arthro.2011.06.026 20. Lyman S, Sherman S, Carter TI, Bach PB, Mandl LA, Marx RG. Prevalence and risk factors for symptomatic thromboembolic events after shoulder arthroplasty. Clin Orthop Relat Res 2006;448:152-6. http://dx.doi.org/10.1097/01.blo.0000194679.87258.6e 21. Navarro RA, Inacio MC, Burke MF, Costouros JG, Yian EH. Risk of thromboembolism in shoulder arthroplasty: effect of implant type and traumatic indication. Clin Orthop Relat Res 2013;471:1576-81. http:// dx.doi.org/10.1007/s11999-013-2829-6 22. Polzhofer GK, Petersen W, Hassenpflug J. Thromboembolic complication after arthroscopic shoulder surgery. Arthroscopy 2003;19:E12932. http://dx.doi.org/10.1016/j.arthro.2003.09.016 23. Randelli P, Castagna A, Cabitza F, Cabitza P, Arrigoni P, Denti M. Infectious and thromboembolic complications of arthroscopic
H. Takahashi et al.
24. 25.
26.
27.
28.
29.
shoulder surgery. J Shoulder Elbow Surg 2010;19:97-101. http://dx. doi.org/10.1016/j.jse.2009.04.009 Scott DL. Pulmonary embolism after elective glenohumeral joint debridement. Orthopedics 2001;24:495-7. Struijk-Mulder MC, Ettema HB, Verheyen CC, B€uller HR. Deep vein thrombosis after arthroscopic anterior cruciate ligament reconstruction: a prospective cohort study of 100 patients. Arthroscopy 2013;29: 1211-6. http://dx.doi.org/10.1016/j.arthro.2013.04.015 Widmer BJ, Bassora R, Warrender WJ, Abboud JA. Thromboembolic events are uncommon after open treatment of proximal humerus fractures using aspirin and compression devices. Clin Orthop Relat Res 2011;469:3332-6. http://dx.doi.org/10.1007/s11999-0111942-7 Willis AA, Warren RF, Craig EV, Adler RS, Cordasco FA, Lyman S, et al. Deep vein thrombosis after reconstructive shoulder arthroplasty: a prospective observational study. J Shoulder Elbow Surg 2009;18: 100-6. http://dx.doi.org/10.1016/j.jse.2008.07.011 Xing KH, Morrison G, Lim W, Douketis J, Odueyungbo A, Crowther M. Has the incidence of deep vein thrombosis in patients undergoing total hip/knee arthroplasty changed over time? A systematic review of randomized controlled trials. Thromb Res 2008;123: 24-34. http://dx.doi.org/10.1016/j.thromres.2008.05.005 Zierler BK. Ultrasonography and diagnosis of venous thromboembolism. Circulation 2004;109:I9-14. http://dx.doi.org/10.1161/01.CIR. 0000122870.22669.4a
www.elsevier.com/locate/ymse
JAPAN SHOULDER SOCIETY ARTICLES
Venous thromboembolism after elective shoulder surgery: a prospective cohort study of 175 patients Hiroyuki Takahashi, MDa, Nobuyuki Yamamoto, MD, PhDa, Hideaki Nagamoto, MDa, Hirotaka Sano, MD, PhDa, Minoru Tanaka, MDb, Eiji Itoi, MD, PhDa,* a b
Department of Orthopaedic Surgery, Tohoku University School of Medicine, Sendai, Japan Department of Orthopaedic Surgery, Tohoku Rosai Hospital, Sendai, Japan Background: The purpose of this study was to investigate the incidence of venous thromboembolism (VTE) after elective arthroscopic shoulder surgery. Materials and methods: One hundred seventy-five consecutive patients who underwent arthroscopic shoulder surgery were enrolled (mean age, 61 years). Patients who had VTE preoperatively and underwent trauma surgery or arthroplasty were excluded. All the patients used foot pumps or elastic stockings after surgery for deep venous thrombosis (DVT) prophylaxis. DVT in the 4 limbs was assessed by ultrasound before and after surgery. Pulmonary embolism was diagnosed by computed tomography pulmonary angiography. Risk factors related to DVT were assessed. Results: The overall incidence of DVT was 10 of 175 patients (5.7%). Most of the DVT cases were detected at 1 to 2 days after surgery. All patients were asymptomatic. There were no patients who had symptomatic pulmonary embolism. However, an asymptomatic pulmonary embolus developed in 1 patient during the 3-month follow-up period. There were no significant differences between the DVT and nonDVT groups regarding the risk factors. Discussion and conclusion: Our data have shown that symptomatic VTE is rare after elective arthroscopic shoulder surgery. However, asymptomatic VTE may occur even with DVT preventive measures. Because most of the DVTs were found in the calf veins, we recommend that surgeons pay attention to the possibility of DVT in the lower extremities even after arthroscopic shoulder surgery. The incidence of asymptomatic VTE after elective arthroscopic shoulder surgery was 5.7%. All patients were asymptomatic, and most of the DVTs occurred 1 or 2 days after surgery. Level of evidence: Level I, Prospective Cohort, Prognosis Study. Ó 2014 Journal of Shoulder and Elbow Surgery Board of Trustees. Keywords: Venous thromboembolism; deep venous thrombosis; pulmonary embolism; shoulder surgery; arthroscopy; asymptomatic; ultrasonography
This study was approved by the institutional review board of Tohoku University Graduate School of Medicine (study No. 2011-250, 2012-1-96). *Reprint requests: Eiji Itoi, MD, PhD, Department of Orthopaedic Surgery, Tohoku University School of Medicine, 1-1, Seiryo-machi, Aoba-ku, Sendai, Japan 980-8574. E-mail address: [email protected] (E. Itoi).
The incidence of calf deep venous thrombosis (DVT) after total knee or hip arthroplasty has been reported to be 20% to 40%.17 Guidelines for the diagnosis, treatment, and prevention of venous thromboembolism (VTE) after total knee or hip arthroplasty and open surgery for
1058-2746/$ - see front matter Ó 2014 Journal of Shoulder and Elbow Surgery Board of Trustees. http://dx.doi.org/10.1016/j.jse.2014.01.054
606 proximal femoral fracture have been published. The guidelines include DVT standard prophylaxis such as intermittent pneumatic compression devices, elastic stockings, and anticoagulation therapy for perioperative management.11,15,17,18 According to the DVT guidelines published by the Japanese Circulation Society in 2009, patients undergoing surgery of the upper extremity are not at risk for postoperative DVT if there are no risk factors related to DVT.17 Preventive measuresdwith the exception of early ambulation and active calf stretch exercisesdare not recommended.11,18 In 1990, Burkhart5 reported the first case of symptomatic DVT after shoulder arthroscopic surgery. Meanwhile, the number of reports about symptomatic VTE after shoulder surgery has been gradually increasing.1,3,4,7,8,10,12-14,16,19-24,27 In 2009, Willis et al27 reported that the prevalence of DVT after shoulder arthroplasty was 13% in 100 patients. Although there has been increased awareness of the existence of DVT after open shoulder surgery, it still has not been a major concern for shoulder surgeons after arthroscopic shoulder surgery. Some retrospective studies have shown that the incidence of symptomatic VTE events after shoulder surgery was extremely low, ranging from 0.01% to 0.68%.8,10,16,20,23 However, there have been only 2 prospective studies investigating the incidence of VTE after elective shoulder surgery.26,27 The purposes of this study were (1) to prospectively investigate the incidence of VTE after elective arthroscopic shoulder surgery and (2) to clarify the risk factors related to VTE events. Our hypothesis was that symptomatic VTE events after elective arthroscopic shoulder surgery are rare whereas asymptomatic VTE might occur even with DVT prophylaxis.
Materials and methods This is a prospective cohort study of VTE that occurred after elective arthroscopic shoulder surgery.
Patients A prospective cohort study was designed to document the incidence of VTE after shoulder surgery in consecutive patients. Between June 2011 and May 2013, a total of 353 consecutive shoulder surgeries were performed at our hospital, as well as a related hospital (one of our affiliated hospitals). Of the patients, 175 who met the following inclusion criteria were enrolled: (1) patients underwent elective arthroscopic shoulder surgery and (2) patients provided informed consent. The exclusion criteria were (1) patients who had a history of symptomatic VTE preoperatively, (2) trauma patients including elective fracture surgery cases, (3) patients who underwent arthroplasty, and (4) patients who were unable to undergo ultrasound examination before and after surgery. Arthroscopic surgery was performed in 175 shoulders in 175 patients. There were 145 arthroscopic rotator cuff surgeries, 17 arthroscopic Bankart repairs, and 13 other
H. Takahashi et al. procedures (Table I). There were 125 men and 50 women, and the mean age at the time of surgery was 61 13 years (range, 1880 years). All procedures were performed with patients under general anesthesia, in the beach-chair position. The procedure was not conducted in the lateral decubitus position in any patient.
DVT preventive measures During and after surgery, DVT prophylaxis such as elastic stockings, pneumatic compression pumps, and early ambulation was used in all patients. Intermittent pneumatic compression devices were used as DVT prophylaxis for the lower extremities during surgery in 173 cases (99%). An elastic stocking for the lower extremities was used in 2 of 175 cases (1%) during surgery. None of the patients in this study used low–molecular weight heparin or aspirin. The pneumatic compression pumps were used from the beginning of surgery for 24 hours. Patients who were not able to begin early ambulation used pneumatic compression pumps until they started ambulation. On the day after surgery, 171 patients (98%) were able to get out of bed. After surgery, intermittent pneumatic compression devices were used in 162 patients (93%), elastic stockings in 2 (1%), and the combination thereof in 11 (6%).
Questionnaire and ultrasound examination A questionnaire was completed before surgery. The potential risk factors associated with VTE, which have been previously reported, were assessed.2,17,27 As preoperative factors, the patient’s background (age, sex, race, body mass index [BMI], and smoking habit), the presence of malignant disease, and comorbidities diagnosed previously were documented (Table II). As perioperative and postoperative factors, operative time, complications, length of hospitalization, and any associated VTE complications, such as symptomatic or fatal pulmonary embolism (PE), were recorded. A 4-limb surveillance duplex ultrasound examination was performed to investigate the incidence of DVT. All examinations were performed and interpreted by a single orthopaedic surgeon specializing in musculoskeletal ultrasonography. In addition, he had 3 months of training in venous ultrasonography to detect DVT in the 4 limbs. To show the accuracy of detecting DVT by ultrasound in this study, the incidence of lower-extremity DVT in patients who underwent total knee arthroplasty during the same period was also investigated by conducting the same ultrasound procedure. Thirty-nine patients who had total knee arthroplasty at our hospital were enrolled, comprising 7 male and 32 female patients with a mean age of 72 years. Their diagnoses were osteoarthritis in 31 cases, rheumatoid arthritis in 7 cases, and psoriatic arthritis in 1 case. The ultrasound examination was performed between 3 days and 1 month after total knee arthroplasty. The ultrasound devices used in this study were the Hi-Vision Preirus system (Hitachi-Aloka Medical, Tokyo, Japan) and MTurbo system (SonoSite, Bothell, WA, USA) with linear probes ranging from 6 to 13 MHz and convex probes ranging from 2 to 5 MHz. The posture was the supine position with the arm in slight abduction. The examination sites were from the elbow joint level to the subclavian vein in the upper extremity and from the posterior calf to the inguinal area in the lower extremity. Duplex
Venous thromboembolism after shoulder surgery Table I
607
Demographic data of all patients
Demographic data No. of patients Age (range) (y) Male-female ratio Diagnosis (n) RCT RAD Others
Overall
Arthroscopic rotator cuff surgery ARCR
A-Patch
ABR
Others
175 61 13 (18-80) 2.5:1
134 61 14 (18-87) 2.5:1
11 69 7 (55-78) 2.7:1
17 38 16 (18-66) 3.3:1
13 60 14 (34-76) 1.6:1
149 (85%) 17 (10%) 9 (5%)
130 (97%) 0 (0%) 4 (3%)
11 (100%) 0 (0%) 0 (0%)
0 (0%) 17 (100%) 0 (0%)
8 (62%) 0 (0%) 5 (38%)
A-Patch, Arthroscopic patch graft procedure; ABR, arthroscopic Bankart repair; ARCR, arthroscopic rotator cuff repair; RAD, recurrent anterior dislocation; RCT, rotator cuff tear.
Table II
Potential VTE risk factors and perioperative data Data
VTE risk factors (n) Age 60 y BMI 30 kg/m2 Comorbidity Diabetes mellitus Hypertension Cardiovascular COPD CKD Chronic liver failure Active cancer Inflammatory disease Smoking habit History/other factors Family history of VTE Cancer Cerebral stroke VTE Major surgery Orthopaedic surgery Trauma (within 6 mo) Additional surgery Perioperative data Operative time (min) Total anesthesia time (min) Length of hospital stay (d)
111 (63%) 7 (4%) 34 (19%) 62 (35%) 14 (8%) 10 (6%) 2 (1%) 3 (2%) 2 (1%) (prostate) 3 (2%) (RA in 2 and Behc¸et disease in 1) 81 (46%) 0 2 8 0 28 33 2 8
(0%) (1%) (5%) (0%) (16%) (19%) (1%) (5%)
135 (95% CI, 126-145) 214 (95% CI, 204-223) 33 (95% CI, 30-35)
CI, Confidence interval; CKD, chronic kidney disease; COPD, chronic obstructive pulmonary disease; RA, rheumatoid arthritis.
ultrasound has been reported to be an accurate and effective screening modality.29 DVT was diagnosed by ultrasound when both of the following findings were present: (1) intravenous hyperechoic lesion or venous incompressibility on B-mode imaging and (2) venous flow defect or abnormality proven in colorflow Doppler mode. When DVT was diagnosed by ultrasound, the serum D-dimer level was also tested to establish the differential diagnosis. Both DVT findings by ultrasound and positive D-dimer
levels indicated a definitive diagnosis of VTE.29 In cases with DVT in which the clinical symptoms of PE were suspected, computed tomography pulmonary angiography (CTPA) was conducted to confirm PE.13 The presence and location of DVT, as well as clinical symptoms related to VTE, were recorded. The mean follow-up period was 10 months after surgery, and DVT after hospital discharge was assessed by the operating surgeon.
Timing of ultrasonographic diagnosis Most thromboembolic events after lower-extremity surgery have been reported to occur in the first postoperative week.28 Because it has been reported that DVT and related thromboembolic complications occur until 12 weeks after knee or hip arthroplasty, it was necessary for us to carefully observe whether VTE events occurred until 3 months after surgery.11,18 Thus, duplex ultrasound was performed (1) before surgery, (2) 1 or 2 days after surgery (acute phase), and (3) 3 weeks to 3 months after surgery (subacute phase). All 175 patients underwent preoperative and postoperative acute-phase examinations, but only 85 patients underwent subacute-phase examinations because some patients or the examiner had time restrictions regarding the ultrasonographic examinations in the outpatient clinic.
Statistical analysis We used c2 analysis or the Fisher exact test to investigate the difference in the incidence of DVT with regard to sex, surgical procedures, smoking habit, and comorbidity. We performed the Mann-Whitney U test to assess the difference in the incidence of DVT with regard to age, BMI, operative time, and other factors. No corrections were made for multiple comparisons. All statistical analyses were performed with JMP Pro 9 software (SAS Institute, Cary, NC, USA). When this study was planned, there was only 1 report available in the literature on the same topic.27 Therefore, it was difficult for us to calculate the estimated sample size. The sample size was determined based on the number of shoulder surgeries that were performed at our hospitals (about 200 cases per year). The power analysis was calculated after this study to check whether the hypothesis was correct. We used G*power 3 software to calculate the statistical power.9
608
H. Takahashi et al.
Figure 1 (A) Ultrasonographic longitudinal image of upper-extremity DVT in a 69-year-old woman, 4 weeks after an arthroscopic patch graft procedure. A fresh free-floating DVT was found in the affected subclavian vein (arrows) on the longitudinal image by ultrasonographic B-mode imaging. (B) Ultrasonographic axial image of upper-extremity DVT. A defect of venous flow was seen in the affected subclavian vein (arrows) on the axial image in ultrasonographic color-flow Doppler mode. SCA, Subclavian artery. (C) Computed tomography pulmonary angiogram. Asymptomatic PE was detected in the right pulmonary artery (arrow).
Results Incidence of DVT The overall incidence of DVT was 10 of 175 patients (5.7%). However, all patients had no symptoms related to VTE. DVT was found in the upper (subclavian vein) and lower (calf vein) extremities in 1 patient and in the lower extremities in 9 patients. DVT was detected in 10 cases at 1 to 2 days after surgery and in 1 case at 3 weeks to 3 months after surgery. The DVT observed in the subclavian vein was a free-floating thrombus (Figs 1, A and B). All DVTs that occurred in the lower extremities were the distal type of DVT. Nine cases had DVTs in the soleus veins, and 1 case had DVT in the gastrocnemius vein. In this study, proximal DVT including popliteal and femoral vein thrombosis was not found. Bilateral calf vein thromboses were detected in 4 cases (Table III). Two patients underwent CTPA. DVTs were found in the bilateral soleus veins on the day after surgery in 1 case. This patient showed a reduction in percutaneous oxygen saturation; this prompted the cardiologist to perform the CTPA examination, without any positive findings.
follow-up. This patient had a DVT in the upper and lower extremities. This asymptomatic and fresh DVT occurred in the subclavian vein of the operative upper extremity at around 4 weeks after an arthroscopic patch graft procedure for a massive rotator cuff tear. Then, an asymptomatic PE was detected by CTPA (Fig 1, C). This patient needed to receive anticoagulant therapy with warfarin for 6 months.
Incidence of VTE after total knee arthroplasty The incidence of lower-extremity DVT in patients who underwent total knee arthroplasty during the same period was 13 of 39 cases (33%). All DVTs were found in the calf veins and were asymptomatic. No symptomatic PE was found.
DVT risk factors The patients whose DVT had been proved by ultrasound (DVT group) and the patients who had no DVT (non-DVT group) were compared. There were no significant differences between the DVT and non-DVT groups regarding the risk factors, such as age, sex, BMI, operative time, smoking habit, and comorbidity (Table IV).
Incidence of PE
Ex post analysis of statistical power
There were no patients who had symptomatic PE in this study. However, an asymptomatic pulmonary embolus developed, as evidenced on CTPA, in 1 patient at 3 months’
Because this study could not predict the risk factors from the incidence of VTE after shoulder surgery, the power analyses after the fact were calculated on the risk factors
Venous thromboembolism after shoulder surgery
609
Table III
Clinical features of VTE patients after shoulder surgery
Age (y)
Sex
Diagnosis
Operative procedure
DVT location
Findings and timing of US (initially postoperatively, at final follow-up)
PE
VTE risk factors
79
M
RCT
ARCR
Bilateral soleus
No
HT
75
M
RCT
ARCR
Bilateral soleus
No
HT
63
M
RCT
ARCR
No
d
58 66
M M
RCT RAD
ARCR ABR
Contralateral gastrocnemius Ipsilateral soleus Contralateral soleus
DVT positive at 1 d, DVT negative at 11 d DVT positive at 2 d, DVT positive at 1 wk DVT positive at 1 d, no follow-up
No No
HT HT
64 70
F M
RCT RCT
ARCR ARCR
Ipsilateral soleus Bilateral soleus
No No
60
M
RCT
ARCR
Ipsilateral soleus
No
HT, DM Prostate cancer d
60
M
RCT
ARCR
Contralateral soleus
No
d
69
F
RCT
A-Patch
Bilateral soleus Affected subclavian
No Yes
HT
DVT positive at 2 d, no follow-up DVT positive at 2 d, DVT negative at 9 d DVT positive at 2 d, no follow-up DVT positive at 1 d, DVT negative at 3 wk DVT positive at 1 d, DVT negative at 3 d DVT positive at 1 d, DVT negative at 3 mo DVT positive at 2 d DVT positive at 4 wk
A-Patch, Arthroscopic patch graft procedure; ABR, arthroscopic Bankart repair; ARCR, arthroscopic rotator cuff repair; DM, diabetes mellitus, F, female; HT, hypertension; M, male; RAD, recurrent anterior dislocation; RCT, rotator cuff tear; US, ultrasonography.
retrospectively. Calculating statistical power for VTE risk factors, we found that the power was in the range between 0.098 and 0.355, which was not sufficient to make any conclusion on the risk factors.
Discussion Although it has been reported that symptomatic VTE after shoulder surgery is a rare complication, there have been no reports describing asymptomatic VTE. Recently, some reports describing VTE after shoulder surgery have been published.1,3,4,7,8,10,12-14,16,19-24,26,27 In addition, it has been pointed out that there is a risk of asymptomatic DVT even after arthroscopic surgery, although the risk of DVT has been reported to be at low.25 However, the risk of VTE after shoulder surgery is very low because there are no highevidence reports about VTE after shoulder surgery according to the guidelines.17 Willis et al27 reported that the prevalence of asymptomatic DVT after shoulder surgery was 13% and the rate of symptomatic PE was 3%. However, all surgeries were arthroplasty cases. In our study, the incidence of asymptomatic DVT after surgery was 5.7% and the proportion of symptomatic PE was 0%. However, asymptomatic PE was detected by CTPA in 1 case. All surgeries in our study were arthroscopic surgeries. The overall incidence of DVT in this study was 5.7% in 175 patients. This difference may come from the different inclusion criteria for the different surgical procedures. Willis et al included patients with a history of VTE events that have been reported as yielding the highest risk of DVT in
the guidelines. All the procedures in their series consisted of shoulder arthroplasties. Our data showed that DVT occurred in all cases at 1 or 2 days after surgery. In contrast to the report of Willis et al, arthroscopic surgeries were performed in all of our cases and a history of VTE was an exclusion criterion in our study. Because of this, the incidence of VTE in this study appears lower than that in the previous report. Only 1 patient underwent anticoagulant treatment with heparin for a few days, because he showed a transient decrease in percutaneous oxygen saturation on monitoring, but this was discontinued after a CTPA showed no PE. Only 1 patient underwent anticoagulant treatment with heparin for a few days until CTPA showed no PE because he showed a transient decrease in percutaneous oxygen saturation on monitoring. In another case, no treatment was performed. Of the 10 cases with DVT, 7 (70%) were followed up by repeat ultrasonography. In 5 of these 7 cases (71%), the DVT findings changed during the follow-up period. Among the cases in which DVT was detected by ultrasound 1 to 2 days after surgery, DVT disappeared between 3 days and 3 months after surgery in 5 of 10 cases (50%) in this study. In the literature, there have been no reports investigating the incidence of DVT longitudinally except for the report of Willis et al. Regarding lower-extremity surgery, Cullison et al6 reported that DVT was detected in only 1 of 67 cases and that asymptomatic DVT was seen in the femoral vein 3 days after surgery but disappeared 10 days after surgery. Willis et al27 conducted ultrasonographic evaluation of 100 patients at 2 days after surgery. Half of these patients (50 patients) were examined at 12 weeks after surgery.
610 Table IV
H. Takahashi et al. Comparison between DVT group and non-DVT group
VTE risk factors
DVT group
Non-DVT group
P value
Power
Mean age (y) Male-female ratio Mean BMI (kg/m2) Mean operative time (min) % with smoking habit % with DVT risk–related comorbidity
66 7 4:1 24.3 3.8 139 (95% CI, 107-171) 70 80
60 14 2.4:1 24.3 3.3 135 (95% CI, 125-145) 45 63
.258 .726 >.99 .507 .191 .334
0.212 0.101 d 0.098 0.355 0.185
CI, Confidence interval.
However, Willis et al did not check for DVTs preoperatively. In their study, because preoperative ultrasound examination was not performed, they cannot completely deny the possibility of pre-existing DVTs. Accordingly, they used the term ‘‘prevalence’’ in their report. In our study, on the other hand, ultrasound examination was performed 3 times: before surgery, 1 to 2 days after surgery (acute phase), and 3 weeks to 3 months after surgery (subacute phase). Thus, we were able to determine the incidence of DVT. These differences may also be related to the difference in the incidences between our study and that of Willis et al. Widmer et al26 reported that the prevalence rates of both DVT and PE after shoulder surgery for proximal humeral fracture were 0%. Their DVT preventive measures were aspirin administration and use of pneumatic compression pumps. The timing of their ultrasonography was, on average, 14 days after surgery, which was earlier than that in our study and in the study by Willis et al.27 The possibility of not being able to detect early DVT after surgery could be a major concern in the study by Widmer et al. They excluded patients who had active cancer and a hypercoagulable state, which have been known to yield a high risk of postoperative VTE. Their finding of a 0% prevalence rate might be caused by the late timing of ultrasonography, the inclusion criteria, and the small sample size of 50 patients. It remains unclear when DVT after surgery occurs in the subacute phase. Therefore, our detection of DVT may have been benefited by defining the subacute phase with a wide range, from 3 weeks to 3 months after surgery in this study. Jameson et al16 reported a large-scale analysis comprising 80,227 patients from a national database. They showed that the overall prevalence of symptomatic VTE events after open shoulder surgery was less than 0.7% and the overall prevalence after arthroscopic procedure was 0.01%. Kuremsky et al19 investigated 1,908 cases of shoulder arthroscopic surgery retrospectively. They found that 6 patients (0.31%) were diagnosed with symptomatic VTE by ultrasound and computed tomography. There have been several reports describing the rarity of symptomatic VTE events after shoulder arthroscopic procedure, and the incidences ranged from 0.01% to 0.38%.3,4,10,14,16,20,23 The incidence of VTE after arthroscopic surgery in our study was 10 of 175 cases (5.7%). This value is higher than the
values in the previous reports, although all VTEs were asymptomatic in our study. To validate the accuracy of ultrasonographic diagnosis for DVT in this study, the incidence of lower-extremity DVT in patients who underwent total knee arthroplasty during the same period was investigated. DVTs were found in 13 of 39 patients (33%), and all were asymptomatic. It has been reported that the incidence of DVT after total knee arthroplasty was 36% on average (range, 25%-59%).28 Because the incidence of DVT in this study was close to our incidence, we assumed that the accuracy of diagnosis of DVT in this study was validated to be at an acceptable level. The surgical position and the anatomic structure of the calf veins have been thought to be a possible cause of DVT formation after shoulder surgery. In our study, soleus vein thromboses were found in 9 of 10 cases. The anatomic and physiological features of the soleus veins are that they are venous sinuses that have a plenty of bypass to the surrounding veins such as gastrocnemius and posterior tibial veins, being predisposed to excessive venous congestion in the lower extremity. The calf vein has been reported as the most common site for DVT observed after lower-extremity surgery.28 With patients in the beach-chair position, the hip and knee joints were always flexed during surgery. This may induce DVT formation in the lower extremities during surgery. Because no surgeries were performed with patients in the lateral decubitus position in our series, the difference between these positions remains unclear. Further studies are necessary to show the difference between these positions. All 10 cases of DVT in our study were detected at 1 to 2 days after surgery. This is similar to the report of Willis et al.27 From this result, we are able to conclude that we need to carefully observe whether VTE events occur soon after shoulder surgery. Considering the fact that most DVTs occur in the calf veins, surgeons should be focused on preventive care for calf DVTs after shoulder surgery regardless of whether VTE-related symptoms are present. There are several limitations in this study. First, the lateral decubitus position was not included. Thus, the incidences of VTE could not be compared between the beach-chair and lateral decubitus positions. Second, DVT prophylaxis during and after surgery was not controlled: elastic stockings, pneumatic compression pumps, and early ambulation were used as
Venous thromboembolism after shoulder surgery DVT prophylaxis. However, intermittent pneumatic compression devices were used in 99% of cases during surgery and in 99% of cases after surgery. Thus, we believe that the effect of variable prophylaxis was small. Third, ultrasonography was performed in only 85 patients at 3 weeks to 3 months after surgery, which was approximately half of all the patients. When we divided the patients into 2 groupsdthat is, the single ultrasonography group, comprising the 90 patients who underwent ultrasound at only 1 to 2 days after surgery, and the followup ultrasonography group, comprising the 85 patients who were able to undergo both acute- and subacute-phase examinations postoperativelydthere were significant differences between the 2 groups in age, diagnosis, operative time, and length of hospital stay. Therefore, it was a limitation of this study not to be able to perform ultrasonographic examination at the same time points. Fourth, ultrasound examination for patients who had total knee arthroplasty was performed at 3 days after surgery or later. Because closed drainage with a compressive bandage was routinely used until 2 days after surgery, ultrasound examination could not be performed during the early postoperative period. Lastly, the ex post analysis of the statistical power for DVT risk factors showed that all power values were lower than 0.4, which was considered insufficient to detect any risk factor in this study. Further prospective studies including greater numbers of patients are needed to determine the risk factors for DVT after elective shoulder surgery.
Conclusions The incidence of VTE after elective arthroscopic shoulder surgery was 5.7%. All patients who had DVTs were asymptomatic, and most of the DVTs occurred 1 or 2 days after surgery.
Acknowledgment We thank Professor Takuhiro Yamaguchi, United Centers for Advanced Research and Translational Medicine, Tohoku University Graduate School of Medicine, for his assistance with the statistical analyses. We appreciate the support of Dr Katsumi Sato, Director, and Dr Ko Himori, Tohoku Rosai Hospital. Our grateful thanks go to the staff of Tohoku University Hospital and Tohoku Rosai Hospital for their assistance in collecting and processing data for the study.
Disclaimer The authors, their immediate families, and any research foundations with which they are affiliated have not received any financial payments or other benefits from
611 any commercial entity related to the subject of this article.
References 1. Amarasekera SS, van Dalen J, Thompson TJ, Osman M. Pulmonary embolism after acromioplasty and rotator cuff repair. J Shoulder Elbow Surg 2008;17:e13-4. http://dx.doi.org/10.1016/j.jse.2008.02.001 2. Anakwe RE, Middleton SD, Beresford-Cleary N, McEachan JE, Talwalkar SC. Preventing venous thromboembolism in elective upper limb surgery. J Shoulder Elbow Surg 2013;22:432-8. http://dx.doi.org/ 10.1016/j.jse.2012.10.033 3. Bongiovanni SL, Ranalletta M, Guala A, Maignon GD. Case reports: heritable thrombophilia associated with deep venous thrombosis after shoulder arthroscopy. Clin Orthop Relat Res 2009;467:2196-9. http:// dx.doi.org/10.1007/s11999-009-0895-6 4. Brislin KJ, Field LD, Savoie FH. Complications after arthroscopic rotator cuff repair. Arthroscopy 2007;23:124-8. http://dx.doi.org/10. 1016/j.arthro.2006.09.001 5. Burkhart SS. Deep venous thrombosis after shoulder arthroscopy. Arthroscopy 1990;6:61-3. 6. Cullison TR, Muldoon MP, Gorman JD, Goff WB. The incidence of deep venous thrombosis in anterior cruciate ligament reconstruction. Arthroscopy 1996;12:657-9. 7. Dattani R, Smith CD, Patel VR. The venous thromboembolic complications of shoulder and elbow surgery: a systematic review. Bone Joint J 2013;95-B:70-4. http://dx.doi.org/10.1302/0301-620X.95B1. 29854 8. Farng E, Zingmond D, Krenek L, Soohoo NF. Factors predicting complication rates after primary shoulder arthroplasty. J Shoulder Elbow Surg 2011;20:557-63. http://dx.doi.org/10.1016/j.jse.2010.11. 005 9. Faul F, Erdfelder E, Lang AG, Buchner A. G*Power 3: a flexible statistical power analysis program for the social, behavioral, and biomedical sciences. Behav Res Methods 2007;39:175-91. http://dx. doi.org/10.3758/BF03193146 10. Garofalo R, Notarnicola A, Moretti L, Moretti B, Marini S, Castagna A. Deep vein thromboembolism after arthroscopy of the shoulder: two case reports and a review of the literature. BMC Musculoskelet Disord 2010;11:65. http://dx.doi.org/10.1186/1471-247411-65 11. Geerts WH, Bergqvist D, Pineo GF, Heit JA, Samama CM, Lassen MR, et al. Prevention of venous thromboembolism: American College of Chest Physicians evidence-based clinical practice guidelines (8th edition). Chest 2008;133:381S-453S. http://dx.doi.org/10. 1378/chest.08-0656 12. Greene JW, Deshmukh AJ, Cushner FD. Thromboembolic complications in arthroscopic surgery. Sports Med Arthrosc 2013;21:69-74. http://dx.doi.org/10.1097/JSA.0b013e31828a7e76 13. Hoxie SC, Sperling JW, Cofield RH. Pulmonary embolism after operative treatment of proximal humeral fractures. J Shoulder Elbow Surg 2007;16:782-3. http://dx.doi.org/10.1016/j.jse.2006.12.004 14. Hoxie SC, Sperling JW, Cofield RH. Pulmonary embolism following rotator cuff repair. Int J Shoulder Surg 2008;2:49-51. http://dx.doi.org/ 10.4103/0973-6042.42576 15. Izquierdo R, Voloshin I, Edwards S, Freehill MQ, Stanwood W, Wiater JM, et al. American academy of orthopaedic surgeons clinical practice guideline on: the treatment of glenohumeral joint osteoarthritis. J Bone Joint Surg Am 2011;93:203-5. 16. Jameson SS, James P, Howcroft DW, Serrano-Pedraza I, Rangan A, Reed MR, et al. Venous thromboembolic events are rare after shoulder surgery: analysis of a national database. J Shoulder Elbow Surg 2011; 20:764-70. http://dx.doi.org/10.1016/j.jse.2010.11.034
612 17. JCS Joint Working Group. Guidelines for the diagnosis, treatment and prevention of pulmonary thromboembolism and deep vein thrombosis (JCS 2009). Circ J 2011;75:1258-81. http://dx.doi.org/10.1253/circj. CJ-88-0010 18. Johanson NA, Lachiewicz PF, Lieberman JR, Lotke PA, Parvizi J, Pellegrini V, et al. American academy of orthopaedic surgeons clinical practice guideline on. Prevention of symptomatic pulmonary embolism in patients undergoing total hip or knee arthroplasty. J Bone Joint Surg Am 2009;91:1756-7. http://dx.doi.org/10.2106/JBJS.I.00511 19. Kuremsky MA, Cain EL, Fleischli JE. Thromboembolic phenomena after arthroscopic shoulder surgery. Arthroscopy 2011;27:1614-9. http://dx.doi.org/10.1016/j.arthro.2011.06.026 20. Lyman S, Sherman S, Carter TI, Bach PB, Mandl LA, Marx RG. Prevalence and risk factors for symptomatic thromboembolic events after shoulder arthroplasty. Clin Orthop Relat Res 2006;448:152-6. http://dx.doi.org/10.1097/01.blo.0000194679.87258.6e 21. Navarro RA, Inacio MC, Burke MF, Costouros JG, Yian EH. Risk of thromboembolism in shoulder arthroplasty: effect of implant type and traumatic indication. Clin Orthop Relat Res 2013;471:1576-81. http:// dx.doi.org/10.1007/s11999-013-2829-6 22. Polzhofer GK, Petersen W, Hassenpflug J. Thromboembolic complication after arthroscopic shoulder surgery. Arthroscopy 2003;19:E12932. http://dx.doi.org/10.1016/j.arthro.2003.09.016 23. Randelli P, Castagna A, Cabitza F, Cabitza P, Arrigoni P, Denti M. Infectious and thromboembolic complications of arthroscopic
H. Takahashi et al.
24. 25.
26.
27.
28.
29.
shoulder surgery. J Shoulder Elbow Surg 2010;19:97-101. http://dx. doi.org/10.1016/j.jse.2009.04.009 Scott DL. Pulmonary embolism after elective glenohumeral joint debridement. Orthopedics 2001;24:495-7. Struijk-Mulder MC, Ettema HB, Verheyen CC, B€uller HR. Deep vein thrombosis after arthroscopic anterior cruciate ligament reconstruction: a prospective cohort study of 100 patients. Arthroscopy 2013;29: 1211-6. http://dx.doi.org/10.1016/j.arthro.2013.04.015 Widmer BJ, Bassora R, Warrender WJ, Abboud JA. Thromboembolic events are uncommon after open treatment of proximal humerus fractures using aspirin and compression devices. Clin Orthop Relat Res 2011;469:3332-6. http://dx.doi.org/10.1007/s11999-0111942-7 Willis AA, Warren RF, Craig EV, Adler RS, Cordasco FA, Lyman S, et al. Deep vein thrombosis after reconstructive shoulder arthroplasty: a prospective observational study. J Shoulder Elbow Surg 2009;18: 100-6. http://dx.doi.org/10.1016/j.jse.2008.07.011 Xing KH, Morrison G, Lim W, Douketis J, Odueyungbo A, Crowther M. Has the incidence of deep vein thrombosis in patients undergoing total hip/knee arthroplasty changed over time? A systematic review of randomized controlled trials. Thromb Res 2008;123: 24-34. http://dx.doi.org/10.1016/j.thromres.2008.05.005 Zierler BK. Ultrasonography and diagnosis of venous thromboembolism. Circulation 2004;109:I9-14. http://dx.doi.org/10.1161/01.CIR. 0000122870.22669.4a