Quality of Life After Axillary or Groin Sentinel Lymph Node Biopsy, With or Without Completion Lymph Node Dissection, in Patients With Cutaneous Melanoma

In the period 1995–2003, a total of 242 patients underwent inguinal and/or axillary SLNB at the University Medical Center Groningen as a staging procedure for stage I/II cutaneous melanoma (>1.0 mm). We excluded 46 patients who had died and 69 patients for one of the following reasons: bilateral axillary or inguinal SLNBs, follow-up in a different area as a result of moving house, preexisting functional limitations and/or previous surgery to the relevant extremities, preexisting volume discrepancies in the relevant extremities (e.g., status after stroke), severe comorbidity such as dementia or disseminated disease, patients undergoing palliative therapy, and patients who were receiving treatment for local or locoregional recurrence at the time of the study.

Therefore, 127 patients were suitable candidates to take part in this study. They were approached by telephone by the research physician. Information about the study was given verbally and they were invited to participate. At their next outpatient follow-up visit, 116 patients (response rate 91%) provided written informed consent. These patients underwent physical examination and were asked to fill in the questionnaire at home and return it to us in the self-addressed, stamped envelope provided. All the patients returned the questionnaire. The internal review board of the hospital approved the study.

Patients were admitted to the hospital to undergo the SLNB, reexcision of scar tissue, and possible CLND. The SLNB procedure has been described in detail previously.24 Excision specimens were sent for routine histopathological analysis with hematoxylin-eosin (HE) staining. Specific immunohistochemical staining was performed on HE-negative specimens for the protein S100 and the melanoma-related antigen HMB45. If histopathological examination of the sentinel lymph node revealed metastatic melanoma tissue, then all those patients received CLND. ALND comprised level I–III axillary dissection with preservation of the pectoralis minor muscle. In the case of a tumor-positive SLNB in the groin, superficial and deep groin lymph node dissections were performed, in which the femoroinguinal lymph nodes and the lymph nodes along the iliac artery and vein were excised, together with the obturator lymph nodes. After superficial lymphadenectomy, the sartorius muscle was freed from its attachment to the anterior superior iliac spine, moved in a medial direction, and fixed to the ligament of Poupart. This technique has been described extensively in the past.25 Since 2002, patients with HE-negative and immunohistochemistry-positive sentinel lymph nodes underwent superficial lymphadenectomy alone. If additional positive lymph nodes were found in the resection specimen, deep groin dissection (iliac and obturator nodal dissection) was performed.

Demographic data (sex, age) and clinical data were extracted from the medical files. Complications in the first 30 days postoperatively comprised wound infection, bleeding, seroma, and wound necrosis.

To determine the degree of lymphedema, the volumes of the arms and/or legs were measured by an adapted version of the water displacement technique as described by Kissin et al.26 Each arm or leg was submerged slowly in a large, transparent water-filled cylinder until the level of the axillary fold or inguinal fold was reached. The extremity was then withdrawn from the cylinder and the displaced water refilled in precise quantities and recorded as the volume of the relevant arm or leg. Both healthy and affected extremities were measured. Lymphedema in the groin was classified according to the criteria formulated by Baas (normal 0–6.5%, slight 6.5–20%, moderate 20–40%, severe >40%). Lymphedema in the axilla was classified according to the criteria formulated by Stillwell (normal 0–10%, slight 10–20%, moderate 20–40%, marked 40–80%, severe >80%).25,27

Descriptive statistics were used to analyze the patients. The χ² and t-tests were used to compare the demographic data and disease-related characteristics between the SLNB tumor-negative group and the tumor-positive group. Unpaired t-tests were computed to compare the patients to the norm group. An analysis of variance (ANOVA) was performed, followed by post hoc Bonferroni tests when the ANOVA was statistically significant, to detect differences between the following four groups: patients with tumor-negative axillary SLNB (axillary SLNB), patients with tumor-positive axillary SLNB followed by ALND (axillary CLND), patients with tumor-negative inguinal SLNB (groin SLNB), and patients with tumor-positive inguinal SLNB followed by inguinal lymph node dissection (groin CLND). In addition, the Kruskal-Wallis test was used to deal with the small numbers. Differences with a P value of <0.05 were considered to be statistically significant. SPSS version 14.0 (SPSS, Chicago, IL) was used to perform all the analyses.

The clinical and pathological data on the melanoma patients are summarized in Table 1. Approximately one-third of the respondents were men, median age in the total group at the time of diagnosis was 50 (range, 18–77) years, median Breslow thickness was 2.0 (range, 1–13) mm, and median follow-up was 56.8 (range, 4–94) months. In the total group, 62 patients underwent inguinal SLNB (53%) and 54 axillary SLNB (47%). The mean number of sentinel lymph nodes excised was 2 (SD = 0.8).

Histopathological examination showed that 27 patients (23%) had metastases from the melanoma in the SLNB: 14 in the groin (of whom 3 had a superficial groin dissection only after the change in surgical decision making in 2002) and 13 in the axilla. These tumor-positive patients underwent CLND. The percentages of men and women were the same in the SLNB tumor-positive and tumor-negative groups. However, there were more women (50 of 62, 81%) than men (12 of 62, 19%, P < 0.001) in the inguinal SLNB group, whereas the percentages of men (30 of 54, 56%) and women (24 of 54, 44%) in the axillary SLNB group were similar (P = 0.414). Patients in the tumor-positive group were significantly older than those in the tumor-negative group. ANOVA showed a significant effect of group on age (F = 2.70, P = 0.049). Subsequent Bonferroni testing showed that the patients in the inguinal tumor-positive group were significantly older (mean = 57.5 years, SD = 11.3) than the patients in the inguinal tumor-negative group (mean = 46.3, SD = 13.9, Bonferroni P = 0.036). No significant difference in age was found between the axillary tumor-positive (mean = 49.1 years, SD = 7.9) and axillary tumor-negative groups (mean = 50.2 years, SD = 14.2).

As far as disease characteristics were concerned, no differences were found between the tumor-positive and tumor-negative groups, except for complications and incidence of lymphedema. Postoperative complications, such as wound infection, seroma, wound necrosis, or bleeding, occurred in 18 patients (16%): 9 (17%) of 54 in the axillary group and 9 (15%) of 62 in the inguinal group. Complications were statistically significantly more common in the tumor-positive group than in the tumor-negative group.

Mean scores of the 116 melanoma patients on the EORTC QLQ-C30 subscales were statistically significantly higher than those in a healthy norm population from Germany. The only exception was cognitive functioning, for which the scores were similar (Table 2). Differences between the study group and the norm group in global QoL and emotional functioning were moderately clinically meaningful; in role functioning, they were small but clinically meaningful. However, in physical, cognitive, and social functioning, they were small and negligible. The QoL scores of the three patients who received a superficial groin dissection due to a change in surgical decision making in 2002 fell within the range of the total group.

ANOVA showed a significant effect of group on role functioning (P = 0.021) and tendencies toward physical functioning problems (P = 0.051) and fatigue (P = 0.051, Table 3). The post hoc Bonferroni test showed that the axillary SLNB tumor-positive group had statistically significantly poorer scores on role functioning than the two tumor-negative groups (Table 3). With regard to the clinical relevance of the differences found between groups, it seemed that in the patients who underwent ALND after SLNB, the scores on global QoL, physical functioning, and social functioning were 5–10 points lower than those in the two tumor-negative groups. In addition, the difference in social functioning compared with the inguinal tumor-positive group was between 5 and 10 points. The difference in role functioning between the axillary tumor-positive group and the three other groups was between 10 and 20 points.

The nonparametric Kruskal-Wallis test confirmed the significant differences in physical functioning (P = 0.01) and in role functioning (P = 0.01), but not the difference in fatigue. In addition, the Kruskal-Wallis test showed significant differences in MPQ Pain Rating Index (P = 0.019) and MPQ Number of Words Chosen (P = 0.019) between the groups. The axillary tumor-positive group seemed to have more functioning problems and symptoms than the other groups.

In view of our findings that the tumor-positive and tumor-negative groups differed in age, complications, and lymphedema, combined with the fact that there were more women than men in the inguinal group than in the axillary group, we performed additional analyses to evaluate how these variables affected the outcome measures. Men had significantly higher mean scores on physical functioning (t = 3.1, P = 0.003) and the GARS (t = −3.1, P = 0.003) and a significantly lower score on fatigue (t = −2.6, P = 0.011) than the women. Age seemed to be negatively but weakly correlated with physical functioning (r = −0.29, P = 0.002) and was positively and moderately correlated with the GARS (r = 0.32, P = 0.001). The older the patient, the poorer the level of physical functioning and execution of daily activities.