- Email: [email protected]
Treatment of Lymphatic Complications after Common Femoral Artery Endarterectomy
Clinical Research Treatment of Lymphatic Complications after Common Femoral Artery Endarterectomy Christian Uhl, Hannah G€otzke, Sandra Woronowicz, Thomas Betz, Ingolf T€opel, and Markus Steinbauer, Regensburg, Germany
Background: This study analyzes the outcome of lymphatic complications after a standard vascular procedure. Methods: This is a retrospective study including patients who had a lymphatic complication after endarterectomy and patch of the common femoral artery in our clinic between March 2007 and June 2018. Therapy of choice was selected according to wound situation and amount of lymphatic liquid. If signs of a wound infection occurred, a surgical therapy was performed; in all other cases a nonsurgical treatment (conservative treatment, radiotherapy) was chosen. Results: We performed 977 index operations, a lymphatic complication occurred in 112 cases (11.5%). In 69 cases the lymphatic complication presented as lymphatic fistula (Group 1), in 43 cases as lymphorrhea from the wound (Group 2). Nonsurgical treatment was done in 66 cases (Group 1: 76.8% vs. Group 2: 30.2%; P < 0.000), and a surgical treatment was necessary in 46 cases (Group 1: 23.2% vs. Group 2: 69.8%; P < 0.000). Indication for surgery was Szilagyi 1 infection in 25 cases, Szilagyi 2 infection in 11 cases, and Szilagyi 3 infection in 10 cases. Patients with Szilagyi 1 infections received negative wound pressure therapy (NWPT). A muscle flap in combination with an NWPT was performed in patients with Szilagyi 2 infections. In Szilagyi 3 infections, the patch was replaced; additionally, a muscle flap and an NWPT were performed. The median hospital stay was 13 days in the nonsurgical group and 22.5 days in the surgical group. We had no bleeding complications and no reinfection during follow-up. The median observation period was 23.0 months. Age 80 years was associated with an increased risk for lymphatic complications. Conclusions: The therapy of lymphatic complications should be done in accordance with clinical symptoms. A nonsurgical treatment is often sufficient. However, in cases of a wound infection different surgical treatments are necessary.
INTRODUCTION The exposure of the common femoral artery is a standard procedure in vascular surgery. Seemingly minor complications like lymphatic fistulas and Funding: This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors. Department of Vascular Surgery, BHB, Regensburg, Germany. Correspondence to: Christian Uhl, MD, Department of Vascular Surgery, Krankenhaus Barmherzige Br€ uder Regensburg, Pr€ ufeninger Straße 86, 93049 Regensburg, Germany; E-mail: [email protected] Ann Vasc Surg 2019; -: 1–5 https://doi.org/10.1016/j.avsg.2019.06.011 Ó 2019 Elsevier Inc. All rights reserved. Manuscript received: May 4, 2019; manuscript accepted: June 2, 2019; published online: - - -
lymphoceles can turn into wound infections with the need of a surgical wound revision. The worst complication that can occur is septic bleeding in the field of the vascular suture. The literature reports lymphatic complications with an incidence of up to 15% following vascular surgery.1,2 However, a standardized definition of lymphatic fistulas does not yet exist.3 Dietl et al. and T€ opel et al. defined a lymphatic fistula as lymphatic secretion of a minimum of 50 mL for more than 4 days and more than 7 days after surgery.4,5 Giovannacci et al.6 defined 30 mL for more than 3 days to be a lymphatic fistula. In other reports, these data are missing. Additionally, lymphoceles can drain through the wound after removal of the drainage. A measurement of the lymphatic liquid is not
1
2 Uhl et al.
possible in these cases. However, these are also postoperative lymphatic complications and should be recorded. Previous reports on postoperative lymphatic complications are very heterogeneous. Several revascularization procedures with different prosthetic or autologous materials are included. Conservative treatment, drainage, negative wound pressure therapy (NWPT), muscle flaps, or radiotherapeutic management are described as therapy opinions. The primary use of fibrin glue during wound closure also seems to have a protective influence.6,7 The aim of this study is to evaluate the outcome of lymphatic complications after a standard vascular surgery procedure. Lymphatic fistulas and lymphorrhea from the wound based on lymphoceles were included in the analysis.
METHODS This is a retrospective analysis of all patients who had lymphatic complications after endarterectomy and patch of the common femoral artery between March 2007 and June 2018 in our clinic. The following was not an exclusion criterion: previous puncture of the femoral artery for cardiac, aortic, iliacal, or peripheral therapy. The study was approved by the local ethics committee. The preparation of the common femoral artery was standardized. We used the lateral approach to avoid a leakage of the lymphatic vessels. If it was necessary to cut lymphatic vessels, we used ligatures. Before arterial clamping patients received 5,000 IU heparin. The endarterectomy included the femoral artery and, when necessary, also the origin of the deep femoral artery and the superficial artery. We exclusively used bovine patches; prosthetic patches were not implanted. The wound was closed by an absorbable subcutaneous suture and a nonabsorbable skin suture. A subcutaneous drainage was inserted to detect postoperative bleeding complications before closing the wound. The drainage was removed if the wound secretion was less than 50 mL/day. All endarterectomies were done under a prophylactic single shot antibiosis. A lymphatic fistula was defined as secretion of 50 mL/day for more than 7 days postoperatively.5 Lymphorrhea from the wound was defined as one or more compresses sodden by lymphatic liquid within 24 hr. If the secretion decreased daily after the operation we did a conservative treatment. Conservative treatment includes limited bed rest and local compression. If the secretion increased we
Annals of Vascular Surgery
started radiotherapy at day 8. Patients received a single dose of 1 Gy/day for 5 consecutive days, in some cases the therapy was extended to at most 10 days. Nonsurgical treatment (conservative treatment or radiotherapy) was only done if there were no signs of wound infection. In cases of a wound infection we performed an operative treatment. The Szilagyi classification (Szilagyi 1e3) was used to assess the severity of the wound infection. A microbial analysis was conducted in all operative cases. In the nonsurgical treated patients we did not do microbial analysis due to no signs of wound infection. Depending on the intraoperative finding we performed an NWPT, a Sartorius muscle flap with additional NWPT, or we replaced the patch by a venous patch and performed an additional muscle flap and NWPT. The patch was replaced in cases of pus around the artery (Szilagyi 3), a muscle flap was performed if the subcutaneous tissue was fragile and a Szilagyi 2 infection existed. An NWPT alone was performed if the arterial vessel was firmly covered by the subcutaneous tissue (Szilagyi 1). Sodden compresses were replaced several times a day when necessary. The NWPT was replaced every fourth day. If the wound secretion ceased, NWPT was removed. The defect of the skin was either closed by skin suture or an open wound healing was established. An antibiotic therapy was started in cases of a positive microbial analysis or in patients with detection of pus. The statistical analysis was done using SPSS 25. Data between the groups were compared using chi-squared test and Fisher’s exact test. A P value <0.05 was considered statistically significant.
RESULTS During this time period we performed 977 endarterectomies of the common femoral artery. Age 80 years was a risk factor for lymphatic complications (P ¼ 0.002). A lymphatic complication occurred in 112 cases (11.5%, 109 patients). Three patients had lymphatic complications at the right and at the left side. In these 3 cases, the endarterectomy was not done simultaneously at both sites. A lymphatic fistula was observed in 69 cases (Group 1) and a lymphocele with secretion from the wound in 43 cases (Group 2). Lymphatic complications and their treatments are shown in Table I. A bovine patch was used in 108 cases and a venous patch out of greater saphenous vein in 4 cases. Indication for the primary use of a venous patch was cardiac intervention within the last 5 days in 3 cases and a fungus in the groin in 1 case. Risk factors and
Volume
-, -
2019
Treatment of lymphatic complications 3
Table I. Lymphatic complications and treatment modalities Lymphatic complications (n ¼ 112)
Nonsurgical treatment Conservative treatment Radiotherapy Surgical treatment NWPT NWPT and muscle flap NWPT, muscle flap, and patch replacement
All (n ¼ 977)
Group 1 (n ¼ 69)
Group 2 (n ¼ 43)
P value
66 (6.8%) 27 (2.8%)
53 (76.8%) 18 (26.1%)
13 (30.2%) 9 (20.9%)
0.000* 0.535
39 46 25 11
35 16 4 6
4 30 21 5
0.000* 0.000* 0.000* 0.421
(4.0%) (4.7%) (2.6%) (1.1%)
10 (1.0%)
(50.7%) (23.2%) (5.8%) (8.9%)
6 (8.9%)
(9.3%) (69.8%) (48.8%) (11.6%)
4 (9.3%)
0.582
*Significant.
indications are shown in Table II. The overall wound complication rate was 12.5%, including lymphatic complications (11.5%), wound break downs without lymphatic source (0.8%), and hematoma (0.2%). A nonsurgical treatment (conservative treatment or radiotherapy) was done in 66 cases (58.9%). From Group 1, 53 cases (76.8%) received this therapy; from Group 2, 13 cases received this therapy (30.2%, P < 0.000). The lymphatic complication occurred in 62 cases during the same hospital stay. In 4 cases, patients were discharged after endarterectomy without pathological findings and returned to the clinic due to lymphatic complications. Of these 66 cases, 27 (24.1%) were treated conservatively and 39 (34.8%) received a radiotherapy (Group 1: 50.7% vs. Group 2: 9.3%; P < 0.000) (Table I). Radiotherapy was done with a median of 5 consecutive days (3e10 days, 1 Gy/time). The success rate of radiotherapy was 78.0%. Initial radiotherapy was started in 50 cases and due to wound impairment a surgical therapy was necessary in 11 cases. We had no late complications related to radiation during follow-up. The median hospital stay between endarterectomy of the common femoral artery and discharge was 12 days for the conservatively treated patients and 17 days in the group that underwent radiotherapy (median 13 days for nonsurgical therapy). We had no further lymphatic complications after discharge in the nonsurgically treated patients. Due to wound impairment 16 cases (23.2%) of Group 1 and 30 cases (69.8%) of Group 2 underwent surgical therapy (P < 0.000). In summary, an operative treatment was performed in 46 cases (41.1%). Of these cases, 17 patients were discharged without pathological findings after endarterectomy and returned to the clinic due to lymphatic complications. The median duration between
endarterectomy and the need for surgical treatment of the lymphatic complication was 13 days. Indication for operation was superficial wound infection in 25 cases (Szilagyi 1, 22.3%) and a deep wound infection without vessel involvement in 11 cases (Szilagyi 2, 9.8%). Furthermore, 10 cases (8.9%) had a Szilagyi 3 infection with pus surrounding the arterial vessel. In 8 of these 10 cases, the bovine patch was removed and a venous patch was performed. In 1 case the common femoral artery was replaced by deep femoral vein and in another case by an Omniflow II graft. A Sartorius muscle flap and an NWPT were additionally performed in these 10 cases. The median hospital stay between endarterectomy and discharge was 24 days in this group. Sartorius muscle flap and NWPT exclusively were done in the cases with a Szilagyi 2 infection. The Table II. Demography, indications, and risk factors All (n ¼ 112)
Age Men Diabetes mellitus Coronary artery disease Myocardial infarction Coronary artery bypass Coronary artery stenting Hypertension Hyperlipidemia Renal insufficiency Dialysis Smoker/ex-smoker Obesity Claudication Rest pain Ulcer/gangrene
71.5 85 43 54 19 19 23 97 69 42 2 39 47 69 24 19
(50e90) (75.9%) (38.4%) (48.2%) (17.0%) (17.0%) (20.5%) (86.6%) (61.6%) (37.5%) (1.8%) (34.8%) (42.0%) (61.6%) (21.4%) (17.0%)
4 Uhl et al.
median time between endarterectomy and hospital discharge was 28 days. An NWPT alone without muscle flap or patch removal was performed in the remaining 25 cases (Group 1: 5.8% vs. Group 2: 48.8%; P < 0.000). The median hospital stay was 20 days (median 22.5 days for surgical therapy). A detection of pathogens was possible in 29 (63.0%) of the 46 operative cases. An antibiotic therapy was established according to the test results. One patient returned to our clinic after NWPT alone with an impairment in the wound again. A vessel involvement was not seen. We performed a successful Sartorius muscle flap in combination with NWPT. Skin sutures were removed 14 days after the last surgical treatment. The in-hospital mortality was 3.5% (4 patients). Two patients died in the radiotherapy group, 1 patient in the surgical group (Sartorius muscle flap and NWPT), and 1 patient in the conservative treatment group. Reason of death was mesenteric ischemia, myocardial infarction, pneumonia, and hepatic failure. We had no bleeding complications and no reinfection. The median follow-up was 23 months (range 1e138, standard deviation ±33.9).
DISCUSSION Lymphatic complications are known side effects after preparation of the femoral vessels. Most of the time lymphorrhea stops after a few days without further therapy. However, in some cases lymphorrhea persists. The risk of postoperative wound infection is increased in patients with lymphatic complications.8,9 Therefore, lymphorrhea therapy is indicated, especially if prosthetic graft material was used for arterial reconstruction. In accordance with a previous paper, we defined a lymphatic fistula as a lymphatic secretion for more than 7 days.5 Giovannacci et al. and Dietl et al. defined a lymphatic fistula as persistent lymphorrhea for more than 3 days, respectively 4 days.4,6 Both authors included patients with prosthetic graft material in their analysis. The concern for graft infections is an explanation for choosing a short period in their definitions. Graft infections are more common in prosthetic materials.10 We used predominantly bovine patches and in some cases autologous vein for a patch. Standardized definitions relating to the amount of lymphatic liquid per day, duration of lymphatic secretion, and therapy concepts are still missing in the literature.7 Our current study shows that not only lymphatic fistulas can result in further complications. Lymphorrhea from the wound also represents a postoperative
Annals of Vascular Surgery
lymphatic complication that should be included. Group 2 included significantly more cases that required surgical treatment. Consequently, we think that maybe the drainage should be left for a longer period of time. However, prospective randomized studies must confirm our hypothesis. Surgical treatment of the lymphatic complication was done at a median of 13 days after endarterectomy. This period between endarterectomy and revision was caused by 2 facts. First, nearly half of the patients of the operatively treated group came back to the hospital after initial discharge due to lymphatic complications. Second, a surgical treatment was performed only in cases of an impairment of the wound. Shermak et al.11 described in their report a median of 14 days as time needed to identify or treat groin wound lymphatic complications. However, it seems that a surgical therapy in lymphatic complications after femoral vessel preparation is not necessary within the first few days. We used different therapy modalities in our study. Conservative therapy, radiotherapy, and surgical therapy were done according to the clinical symptoms. Conservative treatment is the most common initial therapy, lymph leaks can occlude spontaneously in many cases.2 However, standard definition for conservative therapy does not exist and different results are reported. Tyndall et al.1 described a higher infection rate in their conservatively treated group. Kalman et al.12 described conservative and surgical approach as equal effective, drainage volume and duration affects the decision regarding therapy. This is similar to our findings. However, in cases of a wound infection or wound necrosis conservative treatment must be changed to surgical treatment to avoid an infection of the arterial reconstruction. In our study, conservative treatment was successful in 24.1% (27 cases out of 112) of cases. Continuing lymphorrhea is a risk factor for further wound complications. Studies from Germany and Austria described radiotherapy as effective treatment for inguinal lymphorrhea.4,13 So far reports on the side effects do not exist and the risk of tumor induction is insignificant.3 Thus we used this therapy approach in cases in which lymphorrhea did not decrease. In our study, the success rate of lymphatic complications treated by radiotherapy was 78%. This result is similar to other reports.14 However, prospective randomized trials on radiotherapeutic management of lymphatic groin fistulas are missing so far. Additionally, the effectiveness of radiotherapy has not been investigated yet. Maybe therefore radiotherapy is not a common treatment in other countries.
Volume
-, -
2019
Several studies describe the efficiency and safety of NWPT in nonclinical infected and infected fistulas.15e18 Hamed et al.17 described a median duration of 18 days for NWPT therapy. This is similar to our results. However, a study by Dosluoglu et al.16 reported a treatment failure of 17% in Szilagyi 3 infections. We never put NWPT directly on the vessels in order to avoid bleeding from anastomotic sites. If the subcutaneous tissue was fragile or a deep wound infection was detected we performed a muscle flap. Muscle flap surgery is highly effective with a low complication rate in lymphatic groin complications.11 If pathogens were found in the wound, patients received additional antibiotics. Furthermore, the arterial reconstruction was replaced in Szilagyi 3 infections. Current studies reported xenografts to be an effective alternative in septic vascular surgery, especially if autologous materials are not available.19,20 However, a surgical approach was only necessary in approximately 40% of all cases. The duration of in-hospital stay depends on the existence of wound infections and the progress of lymphorrhea. Furthermore, lymphorrhea from the wound seems to require surgical treatment more frequently. This study has several limitations due to the retrospective nonrandomized character and the small cohort. We cannot deduce if radiotherapy is more effective than conservative treatment alone. Prospective studies are necessary to evaluate the effect of radiotherapy in the treatment of lymphatic complications. Furthermore, we have no exact data on time to complete wound closure in cases with open wound healing after removal of the NWPT. Drainage of the lymphocutaneous fistula is also described in the literature for therapy in lymphatic complications.21 However, we did not use this treatment concept in our clinic and therefore we have no data regarding this.
CONCLUSION Due to the varied clinical manifestations and different processes a standard definition for lymphatic complications is difficult. A surveillance period of 7 days before beginning a therapy seems to be justifiable. In many cases a nonsurgical treatment is sufficient. In cases of wound infections a surgical approach is essential. REFERENCES 1. Tyndall SH, Shepard AD, Wilczewski JM, et al. Groin lymphatic complications after arterial reconstruction. J Vasc Surg 1994;19:858e63.
Treatment of lymphatic complications 5
2. Twine CP, Lane IF, Williams IM. Management of lymphatic fistulas after arterial reconstruction in the groin. Ann Vasc Surg 2013;27:1207e15. 3. Habermehl D, Habl G, Eckstein HH, et al. Radiotherapeutic management of lymphatic fistulas. An effective but disregarded therapy option. Chirurg 2017;88:311e6. 4. Dietl B, Pfister K, Aufschlager C, et al. Radiotherapy of inguinal lymphorrhea after vascular surgery. A retrospective analysis. Strahlenther Onkol 2005;181:369e400. 5. T€ opel I, Wiesner M, Uhl C, et al. Retrograde thromboendarterectomy vs. ilio-femoral bypass in unilateral iliac TASC C and D lesions. Vasa 2015;44:211e9. 6. Giovannacci L, Eugster T, Stierli P, et al. Does fibrin glue reduce complications after femoral artery surgery? A randomized trial. Eur J Vasc Endovasc Surg 2002;24:196e201. 7. Giovannacci L, Renggli JC, Eugester T, et al. Reduction of groin lymphatic complications by application of fibrin glue: preliminary results of a randomized study. Ann Vasc Surg 2001;15:182e5. 8. Pedersen G, Laxdal E, Hagala M, et al. Local infections after above-knee prosthetic femoropopliteal bypass for intermittent claudication. Surg Infect (Larchmt) 2004;5:174e9. 9. Illig KA, Alkon JE, Smith A, et al. Rotational muscle flap closure for acute groin wound infections following vascular surgery. Ann Vasc Surg 2004;18:661e8. 10. Siracuse JJ, Nandivada P, Giles KA, et al. Prosthetic graft infections involving the femoral artery. J Vasc Surg 2013;57: 700e5. 11. Shermak MA, Yee K, Wong L, et al. Surgical management of groin lymphatic complications after arterial bypass surgery. Plast Reconstr Surg 2005;115:1954e62. 12. Kalman PG, Walker PM, Johnston KW. Consequences of groin lymphatic fistulae after vascular reconstruction. Vasc Endovasc Surg 1991;25:210e3. 13. Neu B, Gauss G, Haase W, et al. Radiotherapy of lymphatic fistula and lymphocele. Strahlenther Onkol 2000;176:9e15. 14. Mayer R, Simina P, McBride WH, et al. Lymphatic fistulas: obliteration by low-dose radiotherapy. Strahlenther Onkol 2005;181:660e4. 15. Aydin U, Gorur A, Findik O, et al. Therapeutic efficacy of vacuum-assisted-closure therapy in the treatment of lymphatic complications following peripheral vascular interventions and surgeries. Vascular 2015;23:41e6. 16. Dosluoglu HH, Loghmanee C, Lall P, et al. Management of early (<30 day) vascular groin infections using vacuumassisted closure alone without muscle flap coverage in a consecutive patient series. J Vasc Surg 2010;51:1160e6. 17. Hamed O, Muck PE, Smith JM, et al. Use of vacuum-assisted closure (VAC) therapy in treating lymphatic complications after vascular procedures: new approach for lymphoceles. J Vasc Surg 2008;48:1520e3. 18. Greer SE, Adelman M, Kasabian A, et al. The use of subatmospheric pressure dressing therapy to close lymphocutaneous fistulas of the groin. Br J Plast Surg 2000;53:484e7. 19. T€ opel I, Uhl C, Ayx I, et al. Xenografts in septic vascular surgery. Gefasschirurgie 2016;21(Suppl 2):55e8. 20. Lutz B, Reeps C, Biro G, et al. Bovine pericardium as new technical option for in situ reconstruction of aortic graft infection. Ann Vasc Surg 2017;41:118e26. 21. Van den Brande P, von Kemp K, Aerden D, et al. Treatment of lymphocutaneous fistulas after vascular procedures of the lower limb: accurate wound reclosure and 3 weeks of consistent and continuing drainage. Ann Vasc Surg 2012;26:833e8.
Background: This study analyzes the outcome of lymphatic complications after a standard vascular procedure. Methods: This is a retrospective study including patients who had a lymphatic complication after endarterectomy and patch of the common femoral artery in our clinic between March 2007 and June 2018. Therapy of choice was selected according to wound situation and amount of lymphatic liquid. If signs of a wound infection occurred, a surgical therapy was performed; in all other cases a nonsurgical treatment (conservative treatment, radiotherapy) was chosen. Results: We performed 977 index operations, a lymphatic complication occurred in 112 cases (11.5%). In 69 cases the lymphatic complication presented as lymphatic fistula (Group 1), in 43 cases as lymphorrhea from the wound (Group 2). Nonsurgical treatment was done in 66 cases (Group 1: 76.8% vs. Group 2: 30.2%; P < 0.000), and a surgical treatment was necessary in 46 cases (Group 1: 23.2% vs. Group 2: 69.8%; P < 0.000). Indication for surgery was Szilagyi 1 infection in 25 cases, Szilagyi 2 infection in 11 cases, and Szilagyi 3 infection in 10 cases. Patients with Szilagyi 1 infections received negative wound pressure therapy (NWPT). A muscle flap in combination with an NWPT was performed in patients with Szilagyi 2 infections. In Szilagyi 3 infections, the patch was replaced; additionally, a muscle flap and an NWPT were performed. The median hospital stay was 13 days in the nonsurgical group and 22.5 days in the surgical group. We had no bleeding complications and no reinfection during follow-up. The median observation period was 23.0 months. Age 80 years was associated with an increased risk for lymphatic complications. Conclusions: The therapy of lymphatic complications should be done in accordance with clinical symptoms. A nonsurgical treatment is often sufficient. However, in cases of a wound infection different surgical treatments are necessary.
INTRODUCTION The exposure of the common femoral artery is a standard procedure in vascular surgery. Seemingly minor complications like lymphatic fistulas and Funding: This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors. Department of Vascular Surgery, BHB, Regensburg, Germany. Correspondence to: Christian Uhl, MD, Department of Vascular Surgery, Krankenhaus Barmherzige Br€ uder Regensburg, Pr€ ufeninger Straße 86, 93049 Regensburg, Germany; E-mail: [email protected] Ann Vasc Surg 2019; -: 1–5 https://doi.org/10.1016/j.avsg.2019.06.011 Ó 2019 Elsevier Inc. All rights reserved. Manuscript received: May 4, 2019; manuscript accepted: June 2, 2019; published online: - - -
lymphoceles can turn into wound infections with the need of a surgical wound revision. The worst complication that can occur is septic bleeding in the field of the vascular suture. The literature reports lymphatic complications with an incidence of up to 15% following vascular surgery.1,2 However, a standardized definition of lymphatic fistulas does not yet exist.3 Dietl et al. and T€ opel et al. defined a lymphatic fistula as lymphatic secretion of a minimum of 50 mL for more than 4 days and more than 7 days after surgery.4,5 Giovannacci et al.6 defined 30 mL for more than 3 days to be a lymphatic fistula. In other reports, these data are missing. Additionally, lymphoceles can drain through the wound after removal of the drainage. A measurement of the lymphatic liquid is not
1
2 Uhl et al.
possible in these cases. However, these are also postoperative lymphatic complications and should be recorded. Previous reports on postoperative lymphatic complications are very heterogeneous. Several revascularization procedures with different prosthetic or autologous materials are included. Conservative treatment, drainage, negative wound pressure therapy (NWPT), muscle flaps, or radiotherapeutic management are described as therapy opinions. The primary use of fibrin glue during wound closure also seems to have a protective influence.6,7 The aim of this study is to evaluate the outcome of lymphatic complications after a standard vascular surgery procedure. Lymphatic fistulas and lymphorrhea from the wound based on lymphoceles were included in the analysis.
METHODS This is a retrospective analysis of all patients who had lymphatic complications after endarterectomy and patch of the common femoral artery between March 2007 and June 2018 in our clinic. The following was not an exclusion criterion: previous puncture of the femoral artery for cardiac, aortic, iliacal, or peripheral therapy. The study was approved by the local ethics committee. The preparation of the common femoral artery was standardized. We used the lateral approach to avoid a leakage of the lymphatic vessels. If it was necessary to cut lymphatic vessels, we used ligatures. Before arterial clamping patients received 5,000 IU heparin. The endarterectomy included the femoral artery and, when necessary, also the origin of the deep femoral artery and the superficial artery. We exclusively used bovine patches; prosthetic patches were not implanted. The wound was closed by an absorbable subcutaneous suture and a nonabsorbable skin suture. A subcutaneous drainage was inserted to detect postoperative bleeding complications before closing the wound. The drainage was removed if the wound secretion was less than 50 mL/day. All endarterectomies were done under a prophylactic single shot antibiosis. A lymphatic fistula was defined as secretion of 50 mL/day for more than 7 days postoperatively.5 Lymphorrhea from the wound was defined as one or more compresses sodden by lymphatic liquid within 24 hr. If the secretion decreased daily after the operation we did a conservative treatment. Conservative treatment includes limited bed rest and local compression. If the secretion increased we
Annals of Vascular Surgery
started radiotherapy at day 8. Patients received a single dose of 1 Gy/day for 5 consecutive days, in some cases the therapy was extended to at most 10 days. Nonsurgical treatment (conservative treatment or radiotherapy) was only done if there were no signs of wound infection. In cases of a wound infection we performed an operative treatment. The Szilagyi classification (Szilagyi 1e3) was used to assess the severity of the wound infection. A microbial analysis was conducted in all operative cases. In the nonsurgical treated patients we did not do microbial analysis due to no signs of wound infection. Depending on the intraoperative finding we performed an NWPT, a Sartorius muscle flap with additional NWPT, or we replaced the patch by a venous patch and performed an additional muscle flap and NWPT. The patch was replaced in cases of pus around the artery (Szilagyi 3), a muscle flap was performed if the subcutaneous tissue was fragile and a Szilagyi 2 infection existed. An NWPT alone was performed if the arterial vessel was firmly covered by the subcutaneous tissue (Szilagyi 1). Sodden compresses were replaced several times a day when necessary. The NWPT was replaced every fourth day. If the wound secretion ceased, NWPT was removed. The defect of the skin was either closed by skin suture or an open wound healing was established. An antibiotic therapy was started in cases of a positive microbial analysis or in patients with detection of pus. The statistical analysis was done using SPSS 25. Data between the groups were compared using chi-squared test and Fisher’s exact test. A P value <0.05 was considered statistically significant.
RESULTS During this time period we performed 977 endarterectomies of the common femoral artery. Age 80 years was a risk factor for lymphatic complications (P ¼ 0.002). A lymphatic complication occurred in 112 cases (11.5%, 109 patients). Three patients had lymphatic complications at the right and at the left side. In these 3 cases, the endarterectomy was not done simultaneously at both sites. A lymphatic fistula was observed in 69 cases (Group 1) and a lymphocele with secretion from the wound in 43 cases (Group 2). Lymphatic complications and their treatments are shown in Table I. A bovine patch was used in 108 cases and a venous patch out of greater saphenous vein in 4 cases. Indication for the primary use of a venous patch was cardiac intervention within the last 5 days in 3 cases and a fungus in the groin in 1 case. Risk factors and
Volume
-, -
2019
Treatment of lymphatic complications 3
Table I. Lymphatic complications and treatment modalities Lymphatic complications (n ¼ 112)
Nonsurgical treatment Conservative treatment Radiotherapy Surgical treatment NWPT NWPT and muscle flap NWPT, muscle flap, and patch replacement
All (n ¼ 977)
Group 1 (n ¼ 69)
Group 2 (n ¼ 43)
P value
66 (6.8%) 27 (2.8%)
53 (76.8%) 18 (26.1%)
13 (30.2%) 9 (20.9%)
0.000* 0.535
39 46 25 11
35 16 4 6
4 30 21 5
0.000* 0.000* 0.000* 0.421
(4.0%) (4.7%) (2.6%) (1.1%)
10 (1.0%)
(50.7%) (23.2%) (5.8%) (8.9%)
6 (8.9%)
(9.3%) (69.8%) (48.8%) (11.6%)
4 (9.3%)
0.582
*Significant.
indications are shown in Table II. The overall wound complication rate was 12.5%, including lymphatic complications (11.5%), wound break downs without lymphatic source (0.8%), and hematoma (0.2%). A nonsurgical treatment (conservative treatment or radiotherapy) was done in 66 cases (58.9%). From Group 1, 53 cases (76.8%) received this therapy; from Group 2, 13 cases received this therapy (30.2%, P < 0.000). The lymphatic complication occurred in 62 cases during the same hospital stay. In 4 cases, patients were discharged after endarterectomy without pathological findings and returned to the clinic due to lymphatic complications. Of these 66 cases, 27 (24.1%) were treated conservatively and 39 (34.8%) received a radiotherapy (Group 1: 50.7% vs. Group 2: 9.3%; P < 0.000) (Table I). Radiotherapy was done with a median of 5 consecutive days (3e10 days, 1 Gy/time). The success rate of radiotherapy was 78.0%. Initial radiotherapy was started in 50 cases and due to wound impairment a surgical therapy was necessary in 11 cases. We had no late complications related to radiation during follow-up. The median hospital stay between endarterectomy of the common femoral artery and discharge was 12 days for the conservatively treated patients and 17 days in the group that underwent radiotherapy (median 13 days for nonsurgical therapy). We had no further lymphatic complications after discharge in the nonsurgically treated patients. Due to wound impairment 16 cases (23.2%) of Group 1 and 30 cases (69.8%) of Group 2 underwent surgical therapy (P < 0.000). In summary, an operative treatment was performed in 46 cases (41.1%). Of these cases, 17 patients were discharged without pathological findings after endarterectomy and returned to the clinic due to lymphatic complications. The median duration between
endarterectomy and the need for surgical treatment of the lymphatic complication was 13 days. Indication for operation was superficial wound infection in 25 cases (Szilagyi 1, 22.3%) and a deep wound infection without vessel involvement in 11 cases (Szilagyi 2, 9.8%). Furthermore, 10 cases (8.9%) had a Szilagyi 3 infection with pus surrounding the arterial vessel. In 8 of these 10 cases, the bovine patch was removed and a venous patch was performed. In 1 case the common femoral artery was replaced by deep femoral vein and in another case by an Omniflow II graft. A Sartorius muscle flap and an NWPT were additionally performed in these 10 cases. The median hospital stay between endarterectomy and discharge was 24 days in this group. Sartorius muscle flap and NWPT exclusively were done in the cases with a Szilagyi 2 infection. The Table II. Demography, indications, and risk factors All (n ¼ 112)
Age Men Diabetes mellitus Coronary artery disease Myocardial infarction Coronary artery bypass Coronary artery stenting Hypertension Hyperlipidemia Renal insufficiency Dialysis Smoker/ex-smoker Obesity Claudication Rest pain Ulcer/gangrene
71.5 85 43 54 19 19 23 97 69 42 2 39 47 69 24 19
(50e90) (75.9%) (38.4%) (48.2%) (17.0%) (17.0%) (20.5%) (86.6%) (61.6%) (37.5%) (1.8%) (34.8%) (42.0%) (61.6%) (21.4%) (17.0%)
4 Uhl et al.
median time between endarterectomy and hospital discharge was 28 days. An NWPT alone without muscle flap or patch removal was performed in the remaining 25 cases (Group 1: 5.8% vs. Group 2: 48.8%; P < 0.000). The median hospital stay was 20 days (median 22.5 days for surgical therapy). A detection of pathogens was possible in 29 (63.0%) of the 46 operative cases. An antibiotic therapy was established according to the test results. One patient returned to our clinic after NWPT alone with an impairment in the wound again. A vessel involvement was not seen. We performed a successful Sartorius muscle flap in combination with NWPT. Skin sutures were removed 14 days after the last surgical treatment. The in-hospital mortality was 3.5% (4 patients). Two patients died in the radiotherapy group, 1 patient in the surgical group (Sartorius muscle flap and NWPT), and 1 patient in the conservative treatment group. Reason of death was mesenteric ischemia, myocardial infarction, pneumonia, and hepatic failure. We had no bleeding complications and no reinfection. The median follow-up was 23 months (range 1e138, standard deviation ±33.9).
DISCUSSION Lymphatic complications are known side effects after preparation of the femoral vessels. Most of the time lymphorrhea stops after a few days without further therapy. However, in some cases lymphorrhea persists. The risk of postoperative wound infection is increased in patients with lymphatic complications.8,9 Therefore, lymphorrhea therapy is indicated, especially if prosthetic graft material was used for arterial reconstruction. In accordance with a previous paper, we defined a lymphatic fistula as a lymphatic secretion for more than 7 days.5 Giovannacci et al. and Dietl et al. defined a lymphatic fistula as persistent lymphorrhea for more than 3 days, respectively 4 days.4,6 Both authors included patients with prosthetic graft material in their analysis. The concern for graft infections is an explanation for choosing a short period in their definitions. Graft infections are more common in prosthetic materials.10 We used predominantly bovine patches and in some cases autologous vein for a patch. Standardized definitions relating to the amount of lymphatic liquid per day, duration of lymphatic secretion, and therapy concepts are still missing in the literature.7 Our current study shows that not only lymphatic fistulas can result in further complications. Lymphorrhea from the wound also represents a postoperative
Annals of Vascular Surgery
lymphatic complication that should be included. Group 2 included significantly more cases that required surgical treatment. Consequently, we think that maybe the drainage should be left for a longer period of time. However, prospective randomized studies must confirm our hypothesis. Surgical treatment of the lymphatic complication was done at a median of 13 days after endarterectomy. This period between endarterectomy and revision was caused by 2 facts. First, nearly half of the patients of the operatively treated group came back to the hospital after initial discharge due to lymphatic complications. Second, a surgical treatment was performed only in cases of an impairment of the wound. Shermak et al.11 described in their report a median of 14 days as time needed to identify or treat groin wound lymphatic complications. However, it seems that a surgical therapy in lymphatic complications after femoral vessel preparation is not necessary within the first few days. We used different therapy modalities in our study. Conservative therapy, radiotherapy, and surgical therapy were done according to the clinical symptoms. Conservative treatment is the most common initial therapy, lymph leaks can occlude spontaneously in many cases.2 However, standard definition for conservative therapy does not exist and different results are reported. Tyndall et al.1 described a higher infection rate in their conservatively treated group. Kalman et al.12 described conservative and surgical approach as equal effective, drainage volume and duration affects the decision regarding therapy. This is similar to our findings. However, in cases of a wound infection or wound necrosis conservative treatment must be changed to surgical treatment to avoid an infection of the arterial reconstruction. In our study, conservative treatment was successful in 24.1% (27 cases out of 112) of cases. Continuing lymphorrhea is a risk factor for further wound complications. Studies from Germany and Austria described radiotherapy as effective treatment for inguinal lymphorrhea.4,13 So far reports on the side effects do not exist and the risk of tumor induction is insignificant.3 Thus we used this therapy approach in cases in which lymphorrhea did not decrease. In our study, the success rate of lymphatic complications treated by radiotherapy was 78%. This result is similar to other reports.14 However, prospective randomized trials on radiotherapeutic management of lymphatic groin fistulas are missing so far. Additionally, the effectiveness of radiotherapy has not been investigated yet. Maybe therefore radiotherapy is not a common treatment in other countries.
Volume
-, -
2019
Several studies describe the efficiency and safety of NWPT in nonclinical infected and infected fistulas.15e18 Hamed et al.17 described a median duration of 18 days for NWPT therapy. This is similar to our results. However, a study by Dosluoglu et al.16 reported a treatment failure of 17% in Szilagyi 3 infections. We never put NWPT directly on the vessels in order to avoid bleeding from anastomotic sites. If the subcutaneous tissue was fragile or a deep wound infection was detected we performed a muscle flap. Muscle flap surgery is highly effective with a low complication rate in lymphatic groin complications.11 If pathogens were found in the wound, patients received additional antibiotics. Furthermore, the arterial reconstruction was replaced in Szilagyi 3 infections. Current studies reported xenografts to be an effective alternative in septic vascular surgery, especially if autologous materials are not available.19,20 However, a surgical approach was only necessary in approximately 40% of all cases. The duration of in-hospital stay depends on the existence of wound infections and the progress of lymphorrhea. Furthermore, lymphorrhea from the wound seems to require surgical treatment more frequently. This study has several limitations due to the retrospective nonrandomized character and the small cohort. We cannot deduce if radiotherapy is more effective than conservative treatment alone. Prospective studies are necessary to evaluate the effect of radiotherapy in the treatment of lymphatic complications. Furthermore, we have no exact data on time to complete wound closure in cases with open wound healing after removal of the NWPT. Drainage of the lymphocutaneous fistula is also described in the literature for therapy in lymphatic complications.21 However, we did not use this treatment concept in our clinic and therefore we have no data regarding this.
CONCLUSION Due to the varied clinical manifestations and different processes a standard definition for lymphatic complications is difficult. A surveillance period of 7 days before beginning a therapy seems to be justifiable. In many cases a nonsurgical treatment is sufficient. In cases of wound infections a surgical approach is essential. REFERENCES 1. Tyndall SH, Shepard AD, Wilczewski JM, et al. Groin lymphatic complications after arterial reconstruction. J Vasc Surg 1994;19:858e63.
Treatment of lymphatic complications 5
2. Twine CP, Lane IF, Williams IM. Management of lymphatic fistulas after arterial reconstruction in the groin. Ann Vasc Surg 2013;27:1207e15. 3. Habermehl D, Habl G, Eckstein HH, et al. Radiotherapeutic management of lymphatic fistulas. An effective but disregarded therapy option. Chirurg 2017;88:311e6. 4. Dietl B, Pfister K, Aufschlager C, et al. Radiotherapy of inguinal lymphorrhea after vascular surgery. A retrospective analysis. Strahlenther Onkol 2005;181:369e400. 5. T€ opel I, Wiesner M, Uhl C, et al. Retrograde thromboendarterectomy vs. ilio-femoral bypass in unilateral iliac TASC C and D lesions. Vasa 2015;44:211e9. 6. Giovannacci L, Eugster T, Stierli P, et al. Does fibrin glue reduce complications after femoral artery surgery? A randomized trial. Eur J Vasc Endovasc Surg 2002;24:196e201. 7. Giovannacci L, Renggli JC, Eugester T, et al. Reduction of groin lymphatic complications by application of fibrin glue: preliminary results of a randomized study. Ann Vasc Surg 2001;15:182e5. 8. Pedersen G, Laxdal E, Hagala M, et al. Local infections after above-knee prosthetic femoropopliteal bypass for intermittent claudication. Surg Infect (Larchmt) 2004;5:174e9. 9. Illig KA, Alkon JE, Smith A, et al. Rotational muscle flap closure for acute groin wound infections following vascular surgery. Ann Vasc Surg 2004;18:661e8. 10. Siracuse JJ, Nandivada P, Giles KA, et al. Prosthetic graft infections involving the femoral artery. J Vasc Surg 2013;57: 700e5. 11. Shermak MA, Yee K, Wong L, et al. Surgical management of groin lymphatic complications after arterial bypass surgery. Plast Reconstr Surg 2005;115:1954e62. 12. Kalman PG, Walker PM, Johnston KW. Consequences of groin lymphatic fistulae after vascular reconstruction. Vasc Endovasc Surg 1991;25:210e3. 13. Neu B, Gauss G, Haase W, et al. Radiotherapy of lymphatic fistula and lymphocele. Strahlenther Onkol 2000;176:9e15. 14. Mayer R, Simina P, McBride WH, et al. Lymphatic fistulas: obliteration by low-dose radiotherapy. Strahlenther Onkol 2005;181:660e4. 15. Aydin U, Gorur A, Findik O, et al. Therapeutic efficacy of vacuum-assisted-closure therapy in the treatment of lymphatic complications following peripheral vascular interventions and surgeries. Vascular 2015;23:41e6. 16. Dosluoglu HH, Loghmanee C, Lall P, et al. Management of early (<30 day) vascular groin infections using vacuumassisted closure alone without muscle flap coverage in a consecutive patient series. J Vasc Surg 2010;51:1160e6. 17. Hamed O, Muck PE, Smith JM, et al. Use of vacuum-assisted closure (VAC) therapy in treating lymphatic complications after vascular procedures: new approach for lymphoceles. J Vasc Surg 2008;48:1520e3. 18. Greer SE, Adelman M, Kasabian A, et al. The use of subatmospheric pressure dressing therapy to close lymphocutaneous fistulas of the groin. Br J Plast Surg 2000;53:484e7. 19. T€ opel I, Uhl C, Ayx I, et al. Xenografts in septic vascular surgery. Gefasschirurgie 2016;21(Suppl 2):55e8. 20. Lutz B, Reeps C, Biro G, et al. Bovine pericardium as new technical option for in situ reconstruction of aortic graft infection. Ann Vasc Surg 2017;41:118e26. 21. Van den Brande P, von Kemp K, Aerden D, et al. Treatment of lymphocutaneous fistulas after vascular procedures of the lower limb: accurate wound reclosure and 3 weeks of consistent and continuing drainage. Ann Vasc Surg 2012;26:833e8.