Abstract
Background:
Inguinal lymph node dissection (ILND) is used for diagnosis and treatment in penile cancer (PC), vulvar cancer (VC), and melanomas draining to the inguinal lymph nodes. However, ILND is often characterized by its morbidity and high wound complication rate. Consequently, we aimed to characterize wound complication rates after ILND.
Methods:
The National Surgical Quality Improvement Program (NSQIP) database was queried for ILND performed from 2005–2018 for melanoma, PC or VC. 30-day wound complications included wound disruption and superficial/deep/organ-space surgical site infection. Multivariable logistic regression was performed with covariates including cancer type, age, ASA≥3, BMI≥30, smoking history, diabetes, operative time and concomitant pelvic lymph node dissection (PLND).
Results:
1,099 patients had an ILND with 92, 115 and 892 performed for PC, VC, and melanoma respectively. Wound complications occurred in 161 (14.6%) patients, including 12 (13.0%), 17(14.8%) and 132 (14.8%) patients with PC, VC and melanoma, respectively. Median length of stay was 1 day (IQR 0–3 days) and median operative time was 152 minutes (IQR 83–192 minutes). Readmission rate was 12.7%. Wound complications were associated with longer operative time per 10 minutes (OR 1.038, 95% CI 1.019–1.056, p<0.001), BMI≥30 (OR 1.976, 95% CI 1.386–2.818, p <0.001) and concomitant PLND (OR 1.561, 95% CI 1.056–2.306, p=0.025).
Conclusions:
Predictors of wound complications after ILND include BMI≥30, longer operative time and concomitant PLND. There have been efforts to decrease ILND complication rates including minimally invasive techniques and modified templates, which are not captured by NSQIP, and such approaches may be especially considered for those with increased complication risks.
Keywords: Inguinal Lymph Node Dissection, Penile Cancer, Vulvar Cancer, Melanoma
Graphical Abstract
Precis
Inguinal lymph node dissection (ILND) complication rates were characterized using the NSQIP database from 2005–2018. ILND was performed in 1,099 patients with a 20.2% complication rate, including a 14.6% wound complication rate. Wound complication was associated with BMI≥30, longer operative time, and concomitant PLND.
Introduction
Inguinal lymph node dissection (ILND) is used for diagnosis and treatment in penile cancer (PC), vulvar cancer (VC), and melanomas draining to the inguinal lymph nodes.(1–3) According to NCCN guidelines, ILND is indicated for penile cancer if there are palpable lymph nodes or if there are high-risk features such as lymphovascular invasion, ≥pT1G3 or T2 disease, or >50% poorly differentiated.(1) For vulvar cancer, there is a risk of >8% lymphatic metastasis for patients with stage IB-II disease and ILND is recommended in patients ≥ stage IB disease or patients with a sentinel lymph node (SLN) biopsy with >2mm of disease.(2) For melanoma, an anatomically complete dissection of involved nodal basins are required.(3)
While oncologic improvement has been associated with inguinal lymph node dissection, the procedure has been demonstrated to have a high morbidity and wound complication rate including skin edge necrosis, wound dehiscence, infection, lymphocele, femoral vessel and nerve injury, deep venous thrombosis and chronic extremity lymphedema.(4) Historically, series have demonstrated high complication rates up to 80–100%.(5) There have been multiple attempts to decrease complication rates including use of dynamic sentinel node biopsies, modified lymph node dissection templates, and laparoscopic and robotic approaches.(4) Using the National Surgical Quality Improvement Program database, we aimed to characterize the complication rates of inguinal lymph node dissection and determine risk factors associated with wound complications. We hypothesized that complication rates would be associated with comorbidities and be relatively lower compared to historic series since modified templates and minimally invasive techniques have more recently been employed.
Methods
Patient Selection
The Participant Use File of the American College of Surgeons National Surgical Quality Improvement Program (NSQIP) database was queried for ILND performed from 2005–2018 for melanoma, PC or VC which were identified by ICD 9 and 10 codes. For melanoma, ICD 9 codes included 172–172.9 and V10.82, and ICD 10 codes included C43-C43.9 and Z82.820. For vulvar cancer, ICD 9 codes included 184.2, 184.4 and 233.2, and ICD 10 codes included C51.0, C51.8–51.9 and D07.1. For penile cancer, ICD 9 codes included 187.2–187.4 and 233.5, and ICD 10 codes included C60.0-C60.2, C60.8 and C60.9. CPT codes for ILND included 38760 for ILND alone and 38765 for ILND with pelvic lymph node dissection (PLND). Of note, a previous NSQIP study by Glarner et al. in 2013 described CPT 38765 as having a “deep ILND”.(6) However, we considered CPT 38765 as an ILND with concomitant PLND since the coding definition includes dissection of inguinal lymph nodes and the excision of “lymph nodes in the pelvic area…including the external iliac, hypogastric and obturator nodes”.(7, 8)
Variables
Baseline patient characteristics were collected which included age, body mass index (BMI), race, history of diabetes, smoking, dyspnea, chronic obstructive pulmonary disease (COPD), congestive heart failure (CHF), hypertension, bleeding disorder, steroid use and American Society of Anesthesiologist (ASA) score. Additionally, operative characteristics and 30-day postoperative complications were collected which included operative time per 10 minutes, reoperation, readmission, mortality, concomitant PLND, length of stay, deep venous thrombosis, pulmonary embolism, sepsis, septic shock, stroke, myocardial infarction, unplanned intubation, acute renal failure, progressive renal failure, wound disruption, superficial incisional surgical site infection (SSI), deep incisional SSI, organ/space SSI, cardiac arrest, pneumonia, urinary tract infection and transfusion within 72 hours of the start of surgery . Of note, documentation for readmission did not start until 2011. Smoking history was defined as smoking within the past year. Superficial incisional SSI, deep incisional SSI, organ/space SSI and wound disruption were grouped together as wound complications.
Statistical Analysis
Chi-squared test was used for nonparametric categorical variables and Mann-Whitney U test was used for nonparametric continuous variables. Univariable and multivariable logistic regression was performed for any complication and wound complications. Additionally, a univariable and multivariable linear regression was performed for length of stay. Four patients had missing data and were excluded from the regression analyses. Variables included cancer type, age, ASA score ≥ 3, BMI≥30, smoking history, diabetes history, operative time, and presence of concomitant PLND. Additionally, transfusion within 72 hours from start of surgery was added as a variable for wound complications, and presence of any postoperative complication was added as a variable for length of stay. P values <0.05 were considered statistically significant. All statistical analysis was performed using Statistical Package for the Social Sciences (SPSS) software, version 28.
Results
ILND was performed on 1,099 patients with 92 (8.4%), 115 (10.4%) and 892 (81.2%) performed for PC, VC, and melanoma respectively. Baseline patient characteristics can be seen in Table 1. Patients with melanoma were less likely to have ASA ≥ 3 compared to patients with PC and VC (37% vs. 70% vs. 57% respectively, p<0.001). Median length of stay was 1 day (IQR 0–2 days) but was longest for patients with PC compared to VC and melanoma (3 vs. 1 vs. 1 day respectively, p<0.001) (Table 2). Median operative time was 152 minutes (IQR 83–192 minutes), but was also longest for patients with PC compared to VC and melanoma (195 vs. 99 vs. 126 minutes respectively, p<0.001). Complications occurred in 222 (20.2%) of patients with wound complications being most common. Wound complications occurred in 161 (14.6%) patients, including 12 (13.0%), 17 (14.8%) and 132 (14.8%) patients with PC, VC and melanoma, respectively. Reoperation occurred in 29 patients (4.2%) and 78 patients (12.7%) were readmitted. There were three (0.273%) 30-day mortalities.
Table 1.
Baseline Characteristics
| Variable | Overall (n= 1099) |
Penile (n=92) |
Vulvar (n=115) | Melanoma (n=892) | p Value |
|---|---|---|---|---|---|
| Age, y, mean | 58.547 | 62.163 | 63.157 | 57.575 | <0.001 |
| BMI ≥ 30, n (%) | 449 (40.855) | 42 (45.652) | 55 (47.826) | 352 (39.462) | 0.142 |
| ASA ≥ 3, n (%) | 460 (41.856) | 64 (69.565) | 65 (56.522) | 331 (37.108) | <0.001 |
| Race, n (%) | <0.001 | ||||
| White | 942 (85.714) | 68 (73.913) | 89 (77.391) | 785 (88.004) | |
| African American | 32 (2.912) | 9 (9.783) | 5 (4.348) | 18 (2.0179) | |
| Asian | 11 (1.001) | 4 (4.348) | 2 (1.739) | 5 (0.561) | |
| Unknown | 114 (10.373) | 11 (11.957) | 19 (16.522) | 84 (9.417) | |
| Diabetes, n (%) | 146 (13.284) | 27 (29.348) | 22 (19.130) | 97 (10.874) | <0.001 |
| Smoking history, n (%) | 205 (18.653) | 17 (18.478) | 30 (26.087) | 158 (17.713) | 0.0949 |
| Dyspnea, n (%) | 0.6502 | ||||
| At rest | 3 (0.273) | 0 (0) | 1 (0.870) | 2 (0.224) | |
| Moderate exertion | 43 (3.913) | 5 (5.435) | 4 (3.478) | 34 (3.812) | |
| No | 1053 (95.814) | 87 (94.565) | 110 (95.652) | 856 (95.964) | |
| COPD, n (%) | 28 (2.548) | 4 (4.348) | 7 (6.0870) | 17 (1.906) | 0.014 |
| CHF, n (%) | 4 (0.364) | 1 (1.087) | 1 (0.870) | 2 (0.224) | 0.27 |
| Hypertension, n (%) | 439 (39.945) | 49 (53.261) | 55 (47.826) | 335 (37.556) | 0.003 |
| Bleeding disorder, n (%) | 24 (2.184) | 2 (2.174) | 4 (3.478) | 18 (2.018) | 0.601 |
| Recent weight loss, n (%) | 4 (0.364) | 0 (0) | 1 (0.870) | 3 (0.336) | 0.558 |
| Steroid use, n (%) | 22 (2.002) | 1 (1.087) | 4 (3.478) | 17 (1.906) | 0.425 |
ASA, American Society of Anesthesiologists
Table 2.
Operative Details and 30-day Postoperative Complications
| Variable | Overall (n= 1099) |
Penile (n=92) |
Vulvar (n=115) | Melanoma (n=892) | p Value |
|---|---|---|---|---|---|
| Operation time, min, median (IQR) | 129 (83–192) | 195 (154–307.5) | 99 (66–155) | 126 (82–188) | <0.001 |
| Concomitant PLND, n (%) | 259 (23.6) | 36 (39.1) | 25 (21.7) | 198 (22.2) | 0.001 |
| Length of stay, d, median (IQR) | 1 (0–3) | 3 (2–5) | 1 (1–2) | 1 (0–2) | <0.001 |
| Readmission, n (%) | 78 (12.745) | 9 (13.235) | 13 (16.25) | 56 (12.069) | 0.580 |
| Reoperation, n (%) | 29 (4.252) | 4 (5.063) | 2 (2.151) | 23 (4.510) | 0.543 |
| Mortality, n (%) | 3 (0.273) | 0 (0) | 0 (0) | 3 (0.336) | 0.705 |
| Any complication, n (%) | 222 (20.2) | 24 (26.1) | 28 (24.3) | 170 (19.1) | 0.140 |
| Wound complication, n (%) | 161 (14.6) | 12 (13.0) | 17 (14.8) | 132 (14.8) | 0.902 |
| Wound disruption | 20 (1.820) | 3 (3.261) | 0 (0) | 17 (1.906) | 0.198 |
| Superficial incisional SSI | 115 (10.464) | 5 (5.435) | 13 (11.304) | 97 (10.874) | 0.255 |
| Deep incisional SSI | 28 (2.548) | 3 (3.261) | 4 (3.478) | 21 (2.354) | 0.696 |
| Organ/space SSI | 8 (0.728) | 2 (2.174) | 1 (0.870) | 5 (0.561) | 0.219 |
| Deep venous thrombosis, n (%) | 8 (0.728) | 3 (3.261) | 1 (0.870) | 4 (0.448) | 0.010 |
| Pulmonary embolism, n (%) | 4 (0.364) | 0 (0) | 1 (0.870) | 3 (0.336) | 0.558 |
| Sepsis, n (%) | 12 (1.092) | 0 (0) | 1 (0.870) | 11 (1.233) | 0.540 |
| Septic shock, n (%) | 2 (0.182) | 0 (0) | 0 (0) | 2 (0.224) | 0.793 |
| Stroke, n (%) | 0 (0) | 0 (0) | 0 (0) | 0 (0) | - |
| Myocardial infarction, n (%) | 1 (0.091) | 0 (0) | 0 (0) | 1 (0.112) | 0.890 |
| Unplanned intubation, n (%) | 1 (0.091) | 0 (0) | 0 (0) | 1 (0.112) | 0.890 |
| Progressive renal insufficiency, n (%) | 2 (0.182) | 1 (1.087) | 1 (0.870) | 0 (0) | 0.012 |
| Acute renal failure, n (%) | 1 (0.091) | 0 (0) | 0 (0) | 1 (0.112) | 0.890 |
| Cardiac arrest, n (%) | 0 (0) | 0 (0) | 0 (0) | 0 (0) | - |
| Pneumonia, n (%) | 1 (0.091) | 0 (0) | 0 (0) | 1 (0.112) | 0.890 |
| Urinary tract infection, n (%) | 12 (1.092) | 2 (16.667) | 2 (1.740) | 8 (0.897) | 0.415 |
| Transfusion, n (%) | 19 (1.729) | 5 (5.435) | 3 (2.609) | 11 (1.233) | 0.010 |
IQR, interquartile range; PLND, pelvic lymph node dissection; SSI, surgical site infection
On multivariable logistic regression analysis, any complication was associated with longer operative time per 10 minutes (OR 1.052, 95% CI 1.036–1.068, p<0.001), BMI ≥ 30 (OR 1.683, 95% CI 1.228–2.305, p<0.001), and smoking history within the past year (OR 1.574, 95% CI 1.064–2.329) (Table 3). Wound complications were associated with longer operative time per 10 minutes (OR 1.038, 95% CI 1.019–1.056, p<0.001), BMI ≥ 30 (OR 1.976, 95% CI 1.386–2.818, p <0.001) and having concomitant PLND (OR 1.561, 95% CI 1.056–2.306, p=0.025) (Table 4). On multivariable linear regression analysis, longer length of stay was associated with penile cancer (Beta 1.395, 95% CI: 0.52–2.27, p =0.002), history of diabetes (Beta 0.765, 95% CI 0.191–1.34, p=0.009), BMI ≥ 30 (Beta 0.455, 95% CI 0.042 – 0.868, p=0.031), operative time per 10 minutes (Beta 0.126, 95% CI 0.105, 0.146, p<0.001), postoperative complication (Beta 1.108, 95% CI 0.638–1.578, p<0.001) and concomitant PLND (Beta 0.619, 95% CI 0.170–1.069, p =0.007) (Table 5).
Table 3.
Logistic Regression of Any Complication Associated with Inguinal Lymph Node Dissection
| Variable | Univariable | Multivariable | ||||
|---|---|---|---|---|---|---|
| OR | 95% CI | p Value | OR | 95% CI | p Value | |
| Age | 1.01 | 1.000–1.020 | 0.057 | 1.01 | 0.998–1.022 | 0.099 |
| ASA ≥ 3 | 1.735 | 1.290–2.333 | <0.001 | 1.329 | 0.945–1.87 | 0.102 |
| BMI ≥ 30 | 1.915 | 1.423–2.577 | <0.001 | 1.683 | 1.228–2.305 | 0.001 |
| Smoking history | 1.265 | 0.880–1.819 | 0.204 | 1.574 | 1.064–2.329 | 0.023 |
| Diabetes history | 1.534 | 1.028–2.289 | 0.036 | 1.049 | 0.67–1.645 | 0.833 |
| Concomitant PLND | 1.676 | 1.210–2.322 | 0.002 | 1.201 | 0.841–1.715 | 0.315 |
| Operative time per 10 minutes | 1.053 | 1.038–1.068 | <0.001 | 1.052 | 1.036–1.068 | <0.001 |
| Vulvar cancer | Ref | Ref | ||||
| Penile cancer | 1.097 | 0.584–2.061 | 0.774 | 0.541 | 0.268–1.093 | 0.087 |
| Melanoma | 0.732 | 0.463–1.156 | 0.181 | 0.735 | 0.452–1.196 | 0.215 |
ASA, American Society of Anesthesiologists; CI, confidence interval; OR, odds ratio; PLND, pelvic lymph node dissection
Table 4.
Logistic Regression of Wound Complications Associated with Inguinal Lymph Node Dissection
| Variable | Univariable | Multivariable | ||||
|---|---|---|---|---|---|---|
| OR | 95% CI | p Value | OR | 95% CI | p Value | |
| Age | 1.009 | 0.998–1.021 | 0.125 | 1.011 | 0.998–1.025 | 0.105 |
| ASA ≥ 3 | 1.679 | 1.2–2.349 | 0.002 | 1.384 | 0.945–2.028 | 0.095 |
| BMI ≥ 30 | 2.17 | 1.546–3.045 | <0.001 | 1.976 | 1.386–2.818 | <0.001 |
| Smoking history | 1.255 | 0.833–1.89 | 0.277 | 1.552 | 0.999–2.412 | 0.051 |
| Diabetes history | 1.313 | 0.828–2.083 | 0.248 | 0.954 | 0.573–1.589 | 0.858 |
| Concomitant PLND | 1.931 | 1.348–2.767 | <0.001 | 1.561 | 1.056–2.306 | 0.025 |
| Operative time per 10 minutes | 1.040 | 1.025–1.055 | <0.001 | 1.038 | 1.019–1.056 | <0.001 |
| Intraoperative/postoperative transfusion | 2.754 | 1.031–7.355 | 0.043 | 1.037 | 0.32–3.367 | 0.951 |
| Vulvar cancer | Ref | Ref | ||||
| Penile cancer | 0.865 | 0.39–1.916 | 0.72 | 0.484 | 0.204–1.152 | 0.101 |
| Melanoma | 1.001 | 0.579–1.73 | 0.996 | 1.089 | 0.61–1.943 | 0.773 |
CI, confidence interval; OR, odds ratio; PLND, pelvic lymph node dissection
Table 5.
Linear Regression of Length of Stay Associated with Inguinal Lymph Node Dissection
|
Variable |
Univariable | Multivariable | ||||
|---|---|---|---|---|---|---|
| Beta | 95% CI | p Value | Beta | 95% CI | p Value | |
| Age | 0.024 | 0.010, 0.038 | 0.001 | 0.011 | −0.002, 0.024 | 0.105 |
| ASA ≥ 3 | 1.346 | 0.927, 1.764 | <0.001 | −0.024 | −0.404, 0.356 | 0.902 |
| BMI ≥ 30 | 0.556 | 0.130, 0.982 | 0.011 | 0.455 | 0.042, 0.868 | 0.031 |
| Smoking history | −0.248 | −0.787, 0.291 | 0.386 | 0.099 | −0.38, 0.577 | 0.685 |
| Diabetes history | 1.718 | 1.107, 2.328 | <0.001 | 0.765 | 0.191, 1.34 | 0.009 |
| Concomitant PLND | 1.749 | 1.265, 2.233 | <0.001 | 0.619 | 0.17, 1.069 | 0.007 |
| Operative time per 10 minutes | 0.161 | 0.142, 0.179 | <0.001 | 0.126 | 0.105, 0.146 | <0.001 |
| Any postoperative complication | 2.085 | 1.577, 2.594 | <0.001 | 1.108 | 0.638, 1.578 | <0.001 |
| Vulvar cancer | Ref | Ref | ||||
| Penile cancer | 3.202 | 2.260, 4.145 | <0.001 | 1.395 | 0.52, 2.27 | 0.002 |
| Melanoma | −0.054 | −0.722, 0.614 | 0.874 | −0.123 | −0.731, 0.484 | 0.691 |
CI, confidence interval; PLND, pelvic lymph node dissection
Discussion
We demonstrate that the complication rate of ILND was 20.2% with most being wound related in 14.6% of patients. Complication rates did not differ significantly between cancer types; however, length of stay and operative time was longer for penile cancer. On multivariable logistic regression analysis, operative time, BMI≥30 and concomitant PLND was associated with wound complications. Meanwhile, occurrence of any complication was associated with operative time, BMI≥30 and smoking history in the past year. Longer length of stay was associated with longer operative time, BMI≥30, diabetes history, concomitant PLND and penile cancer on multivariable linear regression.
The standard, full template ILND boundaries include the inguinal ligament superiorly, sartorius muscle laterally, the adductor muscle medially and the junction of the adductor longus and sartorius inferiorly(1, 9). Additionally, the floor of the dissection is the fascia lata for the superficial dissection and the pectineus for the deep dissection.(10) Due to this extensive dissection of the inguinal region, lymphatics and neurovasculature can be disrupted leading to high complication rates. Historically, overall complication rates have ranged between 80–100%, with wound infections in 14–17%, skin necrosis and wound dehiscence in 45–62%, and lymphedema in 23–50% of cases.(5) More contemporary studies report overall complication rates between 45 and 57%.(11) Interestingly, our results demonstrate a much lower complication rate compared to prior reports. Glarner et al. analyzed wound complication rates after ILND for melanoma using NSQIP from 2005–2010 and similarly demonstrated a low wound complication rate of 14.2%. The authors associated this to NSQIP not capturing ILND-specific complications such as lymphocele, seroma, hematoma and vascular or neural injury.(6) While the database likely does not capture all ILND-specific complications, several other factors, which are not captured well by NSQIP may also account for the lower complication rate. Such factors include use of different surgical techniques, modified ILND template usage and minimally invasive techniques. For instance, Koifman et al. in 2013 demonstrated that their cohort of 170 patients with PC who underwent a modified template open ILND had a complication rate of 10.3%.(11) They attributed their low complication rate to a horizontal incision made above the inguinal ligament that avoided interruption of blood supply to the skin of the inguinal region. They also avoided transposition of the sartorius muscle, which has been associated with wound complications.(11, 12) Additionally, modified ILND templates have been used more commonly for patients with clinically negative groins. Modified templates have a shorter skin incision, limit the field of the inguinal dissection by excluding the area lateral to the femoral artery and caudal to the fossa ovalis, preserving the saphenous vein and eliminating the need to transpose the sartorius muscle.(1) Bouchot et al. performed 118 modified ILND and 58 radical ILND during treatment of treating PC, and demonstrated lower early and late complications for the modified template (6.8% and 3.4%, respectively) compared to a radical ILND (41.4% and 43.1%, respectively). Additionally, Spiess et al. in 2008 demonstrated a 29% complication rate for 52 patients that underwent a superficial ILND that spared dissection posterior to the fascia lata.(5)
Newer series include patients who have had ILND performed by robotic and/or laparoscopic approaches. Singh et al. in 2018 reported a significant difference in major complication rates between 51 patients with PC who underwent robot-assisted ILND versus 100 with PC who had an open ILND (2% vs. 17% respectively, p=0.0067).(13) Postlewait et al. reported on 102 patients with melanoma or genitourinary malignancies who underwent minimally invasive ILND and demonstrated a relatively low rate of skin necrosis and wound dehiscence of 9.5% compared to other series of open ILND which have ranged from 16–53%.(14) As NSQIP does not have the granularity to capture the usage of modified templates and minimally invasive techniques, it is difficult to discern if such factors contribute to the relatively lower complication rates.
We demonstrated that predictors of ILND wound complications include operative time, BMI≥30 and concomitant PLND. Longer operative time has been demonstrated in a systematic review in 2017 by Cheng et al. as a risk factor for surgical site infections in a variety of surgeries. They showed that mean operative time was thirty minutes longer in patients with SSI compared to those without.(15) Possible reasons for this association include increased duration exposure of incision to environment, which increases risk of bacterial contamination, decreased tissue concentration of perioperative antibiotics, especially if not appropriately re-dosed intraoperatively, and increased procedure complexity. While the NSQIP database does not include oncologic or surgical characteristics, longer operative time may represent those undergoing a more extensive ILND. Elevated BMI has previously been demonstrated as a predictor of ILND wound complications.(6, 16, 17) Wound healing is impaired in obesity due to the avascular nature of adipose tissue, which leads to local hypoxia, impaired neutrophil function and bacterial clearance.(18) Smoking can also affect wound healing through a variety of mechanisms including wound ischemia from decreased microperfusion through platelet aggregation and cutaneous vasocontriction from increased catecholamine release.(18, 19) Smoking was associated with development of postoperative complications, but not wound related complications or length of stay.
There have been mixed results regarding PLND as a predictor of complications. In penile cancer, PLND is recommended if there are ≥2 positive inguinal nodes on the ipsilateral ILND site, extranodal extension or ≥4 positive inguinal nodes on both sides.(1) For melanoma, iliac and obturator lymph node dissections are considered for clinically positive inguinofemoral nodes, positive Cloquet’s node or ≥3 positive inguinofemoral nodes.(3) Gopman et al. in 2014 evaluated 327 patient with PC and demonstrated that 40.1% underwent a concomitant PLND, which was significantly associated with wound complications on univariable analysis, but not multivariable.(20) In contrast, Chang et al. in 2010 showed that 56.6% of their ILND for melanoma underwent a concomitant PLND and PLND was not associated with increased complications.(16) As PLND can cause a variety of complications, such as lymphoceles which can become infected, the added procedure could lead to an increased complication rate.(21, 22) Some predictors of complications seen in other studies have also included surgical and oncologic characteristics including bilateral dissection(12), sartorius muscle transposition(12), open ILND(13), pathologic nodal stage(13), size of lymph node(23) and presence of clinically positive lymph node(19). Identification of patients with risk factors allows for opportunity to optimize patients prior to surgery and to consider modified templates or minimally invasive approaches to decrease morbidity.
Interestingly, length of stay was found to be longer for patients with penile cancer despite complication rates not significantly differing between the cancer types. While this could be explained by the difference in comorbidities (e.g., ASA score, diabetes, smoking) and concomitant PLND rates between cancer types, these factors were accounted for in the multivariable analysis. Additionally, operative time was significantly longer for penile cancer patients. Differences in operative time and length of stay could stem from different surgical and oncologic characteristics which are not captured by NSQIP. More extensive disease, bilateral versus unilateral dissection and radical lymphadenectomy rather than a modified template can increase the complexity of the procedure leading to longer operative times and length of stay. For instance, inguinal lymph node dissection in penile and vulvar cancer traditionally is bilateral; however, in melanoma, laterality depends on the location of the lesion.(1–3) Furthermore, vulvar cancer guidelines describe a <3% risk of contralateral metastasis for a negative unilateral inguinal lymph dissection in vulvar cancer, and that unilateral inguinal lymph node dissection is an option.(2) Therefore, melanoma and vulvar cancer patients may be more likely to undergo a unilateral versus bilateral inguinal lymph node dissection. Future studies could better clarify if length of stay and operative time differences for ILND exist between cancer types, and if they are influenced by surgical and oncologic factors.
There are several limitations to our study. First, NSQIP does not capture information regarding surgical approach since procedures are identified by CPT code, and therefore this study cannot determine how ILND extent, different ILND templates or minimally invasive approaches affect complication rates. Second, NSQIP may not capture certain complications seen with ILND such as skin necrosis, lymphocele, lymphedema, seroma or femoral nerve or vessel injury since they are not specific complications that are recorded. Third, oncologic characteristics are not captured by NSQIP, and therefore the stage of disease or lymph node counts cannot be assessed. Fourth, NSQIP does not capture long-term complication rates and only reports on 30-day complications. Post-operatively, drains can be left in for several days or weeks until the daily output is low.(24) Subsequent lymphoceles may still develop after drain removal but may then occur more than 30 days from the procedure.
Conclusion
This contemporary analysis of ILND complications in the NSQIP database demonstrates that wound complications are the most common complication, and that complication rates do not differ between cancer type. Predictors of wound complications after ILND included BMI≥ 30, operative time and concomitant PLND. There have been efforts to decrease ILND complication rates including modified templates and minimally invasive techniques, which are not captured by NSQIP, and such approaches may be especially considered for those with increased complication risks.
Support:
This work is supported by a grant from the National Cancer Institute (P30CA072720).
Abbreviations:
- NSQIP
National Surgical Quality Improvement Program
- ILND
Inguinal Lymph Node Dissection
- PLND
Pelvic Lymph Node Dissection
- ASA
American Society of Anesthiologist
- BMI
Body Mass Index
- CHF
Congestive Heart Failure
- COPD
Chronic Obstructive Pulmonary Disease
- SSI
Surgical Site Infection
- VC
Vulvar cancer
- PC
Penile Cancer
Footnotes
Disclosure Information: Nothing to disclose.
Disclosures outside the scope of this work: Dr Elsamra is a paid consultant for Intuitive Surgical. Dr Ghodoussipour is a paid consultant for Janssen Pharmaceuticals. Dr Singer receives clinical trial support from Astellas and Medivation, is an advisory board member for Merck, Johnson & Johnson, and Vyriad, and is a data safety monitoring board member for Aura Biosciences.
Publisher's Disclaimer: Disclaimer: The American College of Surgeons National Surgical Quality Improvement Program and the hospitals participating in the ACS NSQIP are the source of the data used herein; they have not verified and are not responsible for the statistical validity of the data analysis or the conclusions derived by the authors.
Presented at the American College of Surgeons 108th Annual Clinical Congress, Scientific Forum, San Diego, CA, October 2022.
References
- 1.National Comprehensive Cancer Network. NCCN Clinical Practice Guidelines in Oncology (NCCN Guidelines®): Penile Cancer Version 2. 2022. Available at. [Google Scholar]
- 2.National Comprehensive Cancer Network. NCCN Clinical Practice Guidelines in Oncology (NCCN Guidelines®): Vulvar Cancer (Squamous Cell Carcinoma) Version 1 2022. Available at. [Google Scholar]
- 3.National Comprehensive Cancer Network. NCCN Clinical Practice Guidelines in Oncology (NCCN Guidelines®): Melanoma: Cutaneous, Version 3 2022. Available at. [Google Scholar]
- 4.Nabavizadeh R, Petrinec B, Nabavizadeh B, et al. Inguinal lymph node dissection in the era of minimally invasive surgical technology. Urol Oncol 2020. Aug 25. [DOI] [PubMed] [Google Scholar]
- 5.Spiess PE, Hernandez MS, Pettaway CA. Contemporary inguinal lymph node dissection: minimizing complications. World Journal of Urology 2008 2008/September/02;27(2):205. [DOI] [PubMed] [Google Scholar]
- 6.Glarner CE, Greenblatt DY, Rettammel RJ, et al. Wound complications after inguinal lymph node dissection for melanoma: is ACS NSQIP adequate? Ann Surg Oncol 2013. Jun;20(6):2049–55. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Smith OJ, Lee-Rodgers L, Ross GL. The nomenclature of groin dissection for melanoma – Time to simplify. Journal of Plastic, Reconstructive & Aesthetic Surgery 2015 2015/November/01/;68(11):1588–91. [DOI] [PubMed] [Google Scholar]
- 8.American Academy of Professional Coders. CPT® 38765, Under Radical Lymphadenectomy (Radical Resection of Lymph Nodes) Available at: https://www.aapc.com/codes/cpt-codes/38765#:~:text=CPT%C2%AE%2038765%2C%20Under%20Radical,Radical%20Resection%20of%20Lymph%20Nodes). Accessed 6/23/2022, 2022.
- 9.Lambert J, Williams M. Chapter 66: Inguinal Lymphadenectomy For Penile Carcinoma. In: Keane TE, Graham SD Jr., eds. Glenn’s Urologic Surgery Eighth Edition. Philadelphia, PA: Wolters Kluwer. [Google Scholar]
- 10.Sharma P, Zargar-Shoshtari K, Pow-Sang JM. Chapter 133: Inguinal Lymphadenectomy. In: Smith JA Jr., Howards S, Preminger G, Dmochowski R, eds. Himan’s Atlas of Urologic Surgery Fourth Edition. Philadelphia, PA: Elsevier. [Google Scholar]
- 11.Koifman L, Hampl D, Koifman N, et al. Radical open inguinal lymphadenectomy for penile carcinoma: surgical technique, early complications and late outcomes. J Urol 2013. Dec;190(6):2086–92. [DOI] [PubMed] [Google Scholar]
- 12.Stuiver MM, Djajadiningrat RS, Graafland NM, et al. Early Wound Complications After Inguinal Lymphadenectomy in Penile Cancer: A Historical Cohort Study and Risk-factor Analysis. European Urology 2013 2013/September/01/;64(3):486–92. [DOI] [PubMed] [Google Scholar]
- 13.Singh A, Jaipuria J, Goel A, et al. Comparing Outcomes of Robotic and Open Inguinal Lymph Node Dissection in Patients with Carcinoma of the Penis. The Journal of Urology 2018 2018/June/01/;199(6):1518–25. [DOI] [PubMed] [Google Scholar]
- 14.Postlewait LM, Farley CR, Diller ML, et al. A Minimally Invasive Approach for Inguinal Lymphadenectomy in Melanoma and Genitourinary Malignancy: Long-Term Outcomes in an Attempted Randomized Control Trial. Annals of Surgical Oncology 2017 2017/October/01;24(11):3237–44. [DOI] [PubMed] [Google Scholar]
- 15.Cheng H, Chen BP-H, Soleas IM, et al. Prolonged Operative Duration Increases Risk of Surgical Site Infections: A Systematic Review. Surgical Infections 2017 2017/August/01;18(6):722–35. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Chang SB, Askew RL, Xing Y, et al. Prospective assessment of postoperative complications and associated costs following inguinal lymph node dissection (ILND) in melanoma patients. Ann Surg Oncol 2010. Oct;17(10):2764–72. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Cirik DA, Karalok A, Ureyen I, et al. Early and Late Complications after Inguinofemoral Lymphadenectomy for Vulvar Cancer. Asian Pac J Cancer Prev 2015;16(13):5175–9. [DOI] [PubMed] [Google Scholar]
- 18.Beyene RT, Derryberry SL Jr., Barbul A. The Effect of Comorbidities on Wound Healing. Surg Clin North Am 2020. Aug;100(4):695–705. [DOI] [PubMed] [Google Scholar]
- 19.Beitsch P, Balch C. Operative morbidity and risk factor assessment in melanoma patients undergoing inguinal lymph node dissection. The American Journal of Surgery 1992;164(5):462–6. [DOI] [PubMed] [Google Scholar]
- 20.Gopman JM, Djajadiningrat RS, Baumgarten AS, et al. Predicting postoperative complications of inguinal lymph node dissection for penile cancer in an international multicentre cohort. BJU Int 2015. Aug;116(2):196–201. [DOI] [PubMed] [Google Scholar]
- 21.Andrews JR, Sobol I, Frank I, et al. Treatment Outcomes in Patients With Symptomatic Lymphoceles Following Radical Prostatectomy Depend Upon Size and Presence of Infection. Urology 2020. Sep;143:181–5. [DOI] [PubMed] [Google Scholar]
- 22.Cacciamani GE, Maas M, Nassiri N, et al. Impact of Pelvic Lymph Node Dissection and Its Extent on Perioperative Morbidity in Patients Undergoing Radical Prostatectomy for Prostate Cancer: A Comprehensive Systematic Review and Meta-analysis. European Urology Oncology 2021 2021/April/01/;4(2):134–49. [DOI] [PubMed] [Google Scholar]
- 23.Serpell JW, Carne PWG, Bailey M. Radical lymph node dissection for melanoma. ANZ Journal of Surgery 2003;73(5):294–9. [DOI] [PubMed] [Google Scholar]
- 24.Lavie O, Karmeli R, Mansano R, et al. Treatment of Recurrent Inguinal Lymphocele by Lymphatic Leakage Mapping and Subsequent Ligation of Lymphatic Vessel Endings: A Case Report. Gynecologic Oncology 2002 2002/January/01/;84(1):155–6. [DOI] [PubMed] [Google Scholar]