Laparoscopic sentinel lymph node procedure using a combination of patent blue and radioisotope in women with cervical carcinoma

Abstract

BACKGROUND

The authors evaluated the feasibility of a laparoscopic sentinel lymph node (SN) procedure with combined radioisotopic and patent blue labeling in patients with cervical carcinoma.

METHODS

Thirteen women (median age, 52.5 years) with cervical carcinoma (Stage Ia2 in 1 patient, Stage Ib1 in 10 patients, Stage Ib2 in 1 patient, and Stage IIa in 1 patient) underwent a laparoscopic SN procedure using an endoscopic γ probe after both radioactive isotope and patent blue injections. After the procedure, all patients underwent complete laparoscopic pelvic lymphadenectomy and either laparoscopic radical hysterectomy (eight patients) or the Schauta–Amreich operation (five patients).

RESULTS

SNs (mean, 1.7 SNs per patient; range, 1–3 SNs per patient) were identified in 12 of 13 patients. A median of 10.5 pelvic lymph nodes per patient (range, 4–17 pelvic lymph nodes per patient) were removed. No lymph node involvement was detected in SNs with hematoxylin and eosin staining. Immunohistochemical studies identified four metastatic SNs in two patients, with micrometastases in two SNs from the first patient and isolated tumor cells in two SNs from the second patient. No false-negative SN results were obtained.

CONCLUSIONS

The results of this study suggest that SN detection with a combination of radiocolloid and patent blue is feasible in patients with cervical carcinoma. The combination of laparoscopy and the SN procedure permitted minimally invasive management of early-stage disease. Cancer 2003;97:3003–9. © 2003 American Cancer Society.

DOI 10.1002/cncr.11423

Cervical carcinoma is the leading gynecologic malignancy worldwide, with 471,000 diagnoses and 233,000 deaths per year.1 Early-stage cervical carcinoma can be cured by surgery and/or radiotherapy.2 Surgical treatment consists of radical hysterectomy and pelvic and/or paraaortic lymphadenectomy.

Lymph node status is a major prognostic factor for patients with cervical carcinoma and is a decision criterion for adjuvant therapy. Pelvic lymph node metastases are detected in only 0–17% of women with Stage Ib cervical carcinoma and in only 12–27% of women with Stage IIa cervical carcinoma.3–6 Thus, as many as four in five of these patients derive no benefit from pelvic lymphadenectomy; on the contrary, this procedure may increase morbidity, especially when postoperative radiotherapy is required.2

Efforts have been made to reduce the morbidity of radical hysterectomy without compromising survival by developing laparoscopic or vaginal approaches. Because current imaging techniques cannot determine lymph node status with accuracy,7 the sentinel lymph node (SN) procedure has emerged as an alternative to systematic lymphadenectomy. The term sentinel lymph node—the first lymph node encountered by lymphatic vessels draining a tumor—was coined in 1960 by Gould et al.8 for carcinoma of the parotid gland. The validity of the SN procedure was established subsequently in patients with melanoma,9 vulvar carcinoma,10 and breast carcinoma.11, 12 In contrast, the SN procedure rarely has been evaluated in patients with cervical carcinoma. Previous studies have demonstrated the relevance of laparoscopic SN procedures using either a blue dye or a radiocolloid.13–15 Recently, Levenback et al.16 reported the feasibility of SN detection with both blue dye and a radioisotope during laparotomy. The objective of the current prospective study was to evaluate the feasibility of a laparoscopic SN procedure with a combination of a radioactive isotope and patent blue in patients with early-stage cervical carcinoma.

MATERIALS AND METHODS

From July 2001 to July 2002, 13 consecutive women who were referred to our institution with early-stage cervical carcinoma were included in the study. Inclusion criteria were age 18–75 years and biopsy-confirmed cervical carcinoma. Exclusion criteria were pregnancy and Stage IIb or more advanced disease. All women underwent preoperative blood sampling, chest X-ray examination, and pelvic magnetic resonance imaging (MRI). The disease stage was classified according to the recommendations of the International Federation of Gynecology and Obstetrics.17

Patients with Stage Ia2 or Ib1 disease underwent initial laparoscopic pelvic lymphadenectomy and vaginal or laparoscopic radical hysterectomy. Patients with Stage Ib2 or IIa disease underwent combined external irradiation and brachytherapy with concomitant chemotherapy. Chemotherapy was administered during the first and fourth week of radiation therapy (RT) and consisted of a continuous 5-fluorouracil infusion at a dose of 1000 mg/m² and a cisplatin bolus (15 mg/m²) 1 hour before RT for 4 consecutive days. The pelvic radiation dose was 40 grays (Gy) in 18 fractions over 29 days (upper limit, lumbar spinal segments 4–5). A vaginal booster dose of 20 Gy was given 5–6 weeks after surgery by means of brachytherapy. The neoadjuvant therapy regimens were followed by laparoscopic pelvic lymphadenectomy and radical hysterectomy. In accordance with our institutional protocol, patients underwent paraaortic lymphadenectomy when metastatic pelvic lymph nodes (assessed by intraoperative pathologic examination) or paraaortic SNs were detected.

The protocol was approved by our Institutional Review Board. All patients provided written consent after receiving all relevant information, including potential adverse effects of patent blue, the radioisotope, general anesthesia, and the laparoscopy procedure and the possible need to convert to open surgery.

SN Procedure

Four injections of 0.2 mL (20 MBq each) of unfiltered technetium sulfur colloid (Nanocis; CIS Bio International, Saclay, France) were administered with a 25-gauge spinal needle on the day before surgery in each quadrant of the cervix. Scintigraphic images were obtained 2 hours after the injections and every 30 minutes thereafter until the SN was visualized, using a triple-head γ camera (Irix; Marconi Corporation, Cleveland, OH). Five-minute static, anterior, and lateral projections were acquired with a low-energy/high-resolution collimator and a matrix size of 512 × 512 pixels. When the SN was not visualized on the day of the injection, imaging was repeated the next day, 2 hours before surgery.

Under general anesthesia, the patients were placed in a low lithotomy position. A speculum was placed in the vagina, and patent blue (Bleu Patenté; V; Guerbet Laboratory, Issy les Moulineaux, France) was injected pericervically through a 25-gauge spinal needle at 3 o'clock and 9 o'clock (1 mL per injection). Antimicrobial chemoprophylaxis (cefazoline 2 g intravenously) was administered at the beginning of the operation. Prophylactic subcutaneous heparin was administered the day before surgery and was continued for 10 days. For the laparoscopic procedure, after pneumoperitoneal insufflation using a Veress needle, a 10 mm laparoscope was inserted through an umbilical incision and was connected to a video monitor. Three stab incisions were made in the suprapubic area: one 12 mm incision in the midline (Versaport; Auto Suture Company, Elancourt, France) and one 5.5 mm incision in each iliac fossa. Six instruments were used: unipolar and bipolar electrocautery forceps, scissors, grasping forceps, and a lavage system.

After patent blue injection, the pelvic and lower paraaortic regions were inspected carefully for lymph ducts and specific dye uptake by lymph nodes. Hot pelvic and paraaortic lymph nodes were located by using an endoscopic γ probe (Eurorad, Strasbourg, France) inserted through the 12 mm suprapubic trocar. Hot lymph nodes were sought before opening the peritoneum. The γ probe was angled laterally to avoid detection of residual radioactivity from the injection site.

After locating the SNs, the peritoneum was opened above the external iliac vessels to the round ligament. Each blue and/or hot lymph node was removed separately in endoscopic bags (Endocatch; Auto Suture Company, Elancourt, France). The relative position of each SN to the major pelvic vessels and the count rate were recorded.

Systematic, laparoscopic, bilateral pelvic lymphadenectomy was performed after the SN procedure. All lymph node tissue along the obturator fossa and the external vessels up to the iliac bifurcation was removed and extracted in an endoscopic bag. The absence of residual pelvic or paraaortic radioactivity was verified before the laparoscopic radical hysterectomy (type III) or the Schauta–Amreich operation, as described previously.18, 19 All patients underwent surgery by the same surgeons (E. D. and E. B.).

The duration of the laparoscopic procedure was calculated from the insertion of the Veress needle to the last skin suture. The duration of the SN procedure was calculated from the insertion of the laparoscope until extraction of the last SN from the abdomen.

Histopathology

SNs and other lymph nodes were inspected by a pathologist. Lymph nodes with macroscopic metastases were sectioned. Normal appearing SNs were cut perpendicular to the long axis. All SNs were submitted to intraoperative imprint cytology. Air-dried cytologic smears were obtained by scraping the cut surfaces and stained with a rapid May–Grünwald–Giemsa method. Each half-SN was sectioned at 3 mm intervals. Each 3 mm section was analyzed by 4 additional levels of 150 μm and 4 parallel sections; 1 was used for hematoxylin and eosin (H&E) staining, and H&E negative sections were examined by immunohistochemistry (IHC) with an anticytokeratin antibody cocktail (cytokeratins AE1-AE3; Dako Corporation, Glostrup, Denmark). Other lymph nodes were submitted totally and blocked individually after 3 mm sectioning and H&E staining.

The sizes of lymph node metastases were estimated with an eyepiece micrometer. Micrometastasis was defined as a single focus of metastatic disease per lymph node measuring no more than 2 mm. The presence of single noncohesive tumor cells was recorded. SNs were recorded as positive when they contained macrometastases, micrometastases, or isolated tumor cells.

Analysis of SNs

SNs were recorded as blue-stained and/or hot (if the ex vivo count exceeded three times the background). The false-negative rate was defined as the number of procedures with a negative SN and one or more positive non-SNs divided by the number of procedures with any positive pelvic lymph node.

RESULTS

Patient Characteristics

The median age of the 13 patients was 52.5 years (range, 34–74 years). Demographic data and tumor characteristics are shown in Table 1. Most patients had Stage Ib disease. The median tumor size was 32 mm (range, 12–55 mm).

Table 1. Demographic Data and Tumor Characteristics of 13 Patients

Characteristic	No. of patients (%)
Stage
Ia2	1 (8)
Ib1	10 (76)
Ib2	1 (8)
Iia	1 (8)
Histology
Squamous	9 (69)
Adenocarcinoma	4 (31)
Preoperative conization	3 (23)
Type of surgery
Laparoscopic radical hysterectomy	8 (62)
Schauta–Amreich operation	5 (38)

The median body mass index was 24.3 kg/m² (range, 18.1–31.1 kg/m²). Three patients were nulliparous. Seven patients were postmenopausal, two of whom were on hormone replacement therapy. One patient had a history of abdominal surgery (for traumatic liver damage).

The SN procedure used both patent blue and radioactive colloid for every patient. None of the patients had MRI evidence of lymph node involvement.

Lymphoscintigraphy

Lymphoscintigraphy always was performed the day before surgery and showed 1 or several uptake foci corresponding to SNs in 12 of 13 patients (Fig. 1). The mean number of hot SNs visualized by preoperative lymphoscintigraphy was 1.7 hot SNs per patient (range, 1–3 hot SNs per patient). Four patients had bilateral hot SNs. All SNs were found in the pelvic region. A patient with negative lymphoscintigraphy results had Stage Ib1 disease (tumor size, 15 mm) without preoperative conization or neoadjuvant therapy. All patients found the radiocolloid injection painful, but none requested analgesic drugs.

Figure 1. Preoperative anterior lymphoscintigraphy. Three sentinel lymph nodes (SN) are visible in the pelvis 2 hours after injection of radioactive isotope (arrow).

Laparoscopic SN Procedure

The median interval between patent blue injection and the beginning of the laparoscopic SN procedure was 6 minutes (range, 5–10 minutes). The median duration of the laparoscopic SN procedure was 16 minutes (range, 10–35 minutes).

The SN was identified in 12 patients (92.3%). The patient in whom no SN was identified by lymphoscintigraphy had no blue or hot SN detected by laparoscopy. The mean number of SNs was 1.7 per patient (range, 1–3 SNs per patient). A total of 21 SNs were removed. Of these, 19 SNs were both blue and hot (Fig. 2), 2 SNs were hot alone, and 0 SNs were blue alone.

Figure 2. Intraoperative laparoscopic view of the pelvis. (A) Identification of two blue-dyed lymphatic channels on the right side of the pelvis (arrow). (B) Removal of a blue-dyed and hot sentinel lymph node that was detected by an endoscopic γ probe. The external iliac artery (EIA) and the external iliac vein (EIV) are indicated.

Nine patients had a unilateral SN (on the right side in six patients and on the left side SN in three patients), and three patients had bilateral SNs. The most common site of the SNs was the medial external iliac region (48%) (Table 2). No SNs were identified in parametrial or paraaortic sites. The patient who had a common iliac SN had two other SNs in the ipsilateral external iliac region.

Table 2. Location of 21 Sentinel Lymph Nodes

Site	No. of sentinel lymph nodes (%)
Common iliac	1 (5)
External iliac
Lateral group	1 (5)
Medial group	10 (48)
Obturator fossa	3 (14)
Interiliac area	6 (28)
Total	21 (100)

Two patients received neoadjuvant chemotherapy, external radiotherapy, and brachytherapy before the surgical procedure. These preoperative treatments did not affect the performance of the SN procedure, because two blue and hot SNs were identified in the external iliac region for each patient.

The median number of lymph nodes removed, including SNs, was 10.5 per patient (range, 4–17 lymph nodes removed per patient). None of the patients required paraaortic lymphadenectomy, because no positive SNs were found by intraoperative histologic examination.

The mean total operating time, including the laparoscopic SN procedure, complete lymph node tissue removal, and radical hysterectomy, was 246 minutes (range, 210–300 minutes). No difference in operating time was found between patients who underwent radical laparoscopic hysterectomy and patients who underwent a Shauta–Amreich operation.

No anaphylactic reactions to patent blue were noted. There were no complications of the laparoscopic SN procedure and no laparoconversions.

Histologic Findings

None of the 21 SNs showed signs of malignancy on intraoperative imprint cytology. Therefore, no patient underwent paraaortic lymphadenectomy.

No lymph space involvement was detected in the SNs by H&E examination. IHC studies showed involvement of four SNs in two patients with Stage Ib1 disease, including micrometastases in two SNs from the first patient and isolated tumor cells in two SNs from the second patient. Two of these SNs were located in the obturator region, and the other two were located in the medial external iliac region. The SNs were the only histopathologically positive lymph nodes in these two patients. We found no false-negative SNs.

No further external radiotherapy was prescribed for women with micrometastases or isolated tumor cells. The pelvic lymph nodes were uninvolved in the patient with uninvolved SNs.

DISCUSSION

The current study demonstrates the feasibility of a laparoscopic SN procedure based on combined detection with patent blue and a radiocolloid. The identification rate was 92%. Our results are in keeping with those of Dargent et al.,15 who reported that a laparoscopic SN procedure with patent blue alone identified SNs in 100% of patients. Likewise, Kamprath et al.20 reported an identification rate of 89% with a laparoscopic procedure and radiocolloid alone.20 However, in their study, SN detection was performed extracorporeally after patients underwent systematic, complete lymphadenectomy. Using patent blue alone, radiocolloid alone, or a combination of the 2 reagents, Malur et al.21 reported SN detection rates of 55%, 76%, and 90%, respectively; however, an unspecified number of patients underwent laparoscopic SN procedure. Recently, Levenback et al.16 confirmed the relevance of the SN procedure in a large series using a combined method, but only with laparotomy. Except for the study by Echt et al.,13 in which the SN detection rate was only 15%, the SN yield ranged between 60% and 100%, regardless of the surgical route or the detection method (Table 3). The sensitivity of the SN procedure ranged from 87% to 100% with an accuracy rate of 100% and without specific morbidity.13–16, 22, 23

Table 3. Sentinel Lymph Node Procedures in Patients with Cervical Carcinoma

Reference	No. of patients	FIGO stage	Injection technique	Route for SN procedure	SN detection rate (%)	Mean no. of SNs removed	IHC analysis of SNs	No. of patients with positive SNs	FN (%)
FIGO: International Federation of Gynecology and Obstetrics; SN: sentinel lymph node; IHC: immunohistochemistry; FN: false-negative; B: colorimetric method; R: radioactive method; NS: not stated. a SN detection was performed extracorporeally after systematic complete lymphadenectomy.
Echt et al., 199913	13	Ib	B	Laparotomy	15	1.5	No	2	0
Medl et al., 200030	3	NS	B	Laparotomy	100	NS	NS	3	0
O'Boyle et al., 200014	20	Ib, IIa	B	Laparotomy	60	1.9	No	3	0
Dargent et al., 200015	35	Ia2, Ib	B	Laparoscopy	100	1.8	No	11	0
Verheijen et al., 200022	10	Ib	B, R	Laparotomy	80	2.3	Yes	1	0
Kamprath et al., 200020	18	I, II	Ra	Laparoscopy	89	NS	NS	1	0
Lantzsch et al., 200123	14	Ib1	R	Laparotomy	93	2	Yes	1	0
Malur et al., 200121	50	I, II, IV	B, R	Laparotomy, laparoscopy	78	2	No	5	17
Levenback et al., 200216	39	Ia, Ib, IIa	B, R	Laparotomy	100	4.7	Yes	8	11
Current study	13	Ia, Ib, IIa	B, R	Laparoscopy	92	1.7	Yes	2	0

In the current study, SN detection was possible in patients who had undergone neoadjuvant chemotherapy and radiotherapy. No previous data were available on the SN procedure in women with cervical carcinoma who were treated first with chemotherapy or radiotherapy. This raised the issue of SN procedures in patients who had locally advanced cervical carcinoma, especially because one randomized trial showed increases in disease free survival and overall survival after patients received neoadjuvant chemotherapy and radiotherapy.24 In the same way, in patients with breast carcinoma, the SN procedure was not used initially after neoadjuvant treatment; however, its feasibility has been demonstrated recently.25

In the current series, all SNs were hot, and all but two were blue. However, it should be noted that radiocolloid was injected into four sites, and patent blue was injected into only two sites. In part, our data are in line with those of Malur et al.,21 who reported that the SN procedure using a combined detection method gave a higher SN detection rate compared with blue staining alone. Conversely, some SNs may stain blue but remain cold, providing further justification for the use of a combined detection method.16, 22

All but one of the SNs in the current study were located in the pelvis, in keeping with previous studies (Table 4) showing that SNs are located preferentially in the external iliac region. SN detection rates in the common iliac region range from 0% to 17% (Table 4). In the current study, no parametrium SNs were detected. This may be explained in part by the background related to the proximity of cervical injection sites. No paraaortic SNs were found in our study. In contrast, Levenback et al.16 found paraaortic SNs in 9% of patients, but the presence of paraaortic SNs with or without external or common iliac SNs was not specified.16 In the current study, the patient who had a common iliac SN also had two ipsilateral external iliac SNs. This is in keeping with reports that skipped lymph node drainage is rare in women with cervical carcinoma.14–16, 22, 23 Our findings also support the concept that the main lymph drainage route in cervical carcinoma runs along the caudal, medial, or cephalic surface of the external iliac vein.26 This would explain why the reported rate of paraaortic metastasis without pelvic lymph node involvement is only 0.0–1.2%.3, 4

Table 4. Locations of Sentinel Lymph Nodes in Patients with Cervical Carcinoma

Reference	No. of SNs (%)
	No. of LNs removed	Paraaortic	Common iliac	External iliac					Internal iliac	Parametrial
				Lateral group	Intermediate group	Medial group	Obturator	Interiliac
SNs: sentinel lymph nodes.
Echt et al., 199913	3	0	1	—	—	2	—	—	0	0
O'Boyle et al., 200014	23	0 (0)	4 (17)	6 (27)	0 (0)	7 (31)	1 (4)	0 (0)	1 (4)	4 (17)
Dargent et al., 200015	63	0 (0)	3 (5)	18 (28)	13 (21)	22 (35)	0 (0)	0 (0)	0 (0)	7 (11)
Verheijen et al., 200022	18	0 (0)	3 (17)	—	8 (44)	—	7 (39)	0 (0)	0 (0)	0 (0)
Lantzsch et al., 200123	26	0 (0)	0 (0)	—	17 (65)	—	3 (12)	0 (0)	5 (19)	0 (0)
Levenback et al., 200216	132	12 (9)	15 (11)	—	32 (24)	—	30 (23)	31 (24)	0 (0)	12 (9)
Current study	21	0 (0)	1 (5)	1 (5)	0 (0)	10 (48)	3 (14)	6 (28)	0 (0)	0 (0)

In the current study, 2 of 13 patients (16.6%) had lymph node metastases. No SN involvement was detected either by imprint cytology or by H&E staining. Immunohistochemical SN examination identified occult metastases in two patients. Previous IHC studies of SN showed no micrometastases.16, 22, 23 Similar to patients with breast carcinoma, the prognostic significance of micrometastases is controversial in patients with cervical carcinoma. At our institution, their presence does not influence the treatment regimen. However, despite the absence of lymph node involvement and clear surgical margins around primary cervical tumors, pelvic recurrences can occur in up to 10% of patients.27

In the current series, no metastatic lymph nodes were found when the SN were negative, in keeping with several previous studies.13–15, 20, 23, 25, 28 In contrast, Levenback et al.16 reported a false-negative result in a patient with a positive medial parametrial lymph node that was resected with the primary tumor and was not located with blue dye or a γ probe. Malur et al.21 observed 1 false-negative result in a series of 50 patients but did not state the location of the positive lymph node.

The numbers of patients in the current study were too small to establish a learning curve for the SN procedure. In patients with breast carcinoma, several authors have reported that at least 20–30 SN procedures, combined with systematic axillary lymphadenectomy, have to be performed before the SN technique can be adopted routinely.29 It is difficult to recommended the same attitude for the pelvic SN procedure, because no consensus on the SN detection technique is available and because the incidence of cervical carcinoma is between 8 times and 15 times lower compared with the incidence of breast carcinoma. In addition, a major issue in the pelvic SN procedure is the risk of lymph node recurrence. Because lymph node metastases are detected in only 0–17% of women with Stage Ib cervical carcinoma and only 12–27% of women with Stage IIa cervical carcinoma,3–6 the detection of a 50% increase in the rate of lymph node recurrence after the SN procedure would require at least 400 women in each arm, with a type I error of 0.05 and a type II error of 0.2.

In conclusion, the current results confirm the feasibility of an SN procedure that, when used in conjunction with laparoscopic radical hysterectomy or the Schauta–Amreich operation, offers a less aggressive management approach for women with early-stage cervical carcinoma. However, the SN procedure must be standardized before it can be recommended for routine use.