Complications associated with iliosacral lymphadenectomy in dogs with metastatic apocrine gland anal sac adenocarcinoma

Abstract

Objective

To report intraoperative and immediate postoperative complications associated with removal of metastatic iliosacral lymph nodes in dogs with apocrine gland anal sac adenocarcinoma.

Animals

There were 136 client-owned dogs in the study.

Procedure

Retrospective multi-institutional study. The database of collaborating institutions was searched for dogs with metastatic apocrine gland anal sac adenocarcinoma that underwent lymphadenectomy for removal of one or more iliosacral lymph nodes. Information of signalment, hematological abnormalities, abdominal computed tomography or ultrasound findings, number and size of enlarged lymph nodes, intraoperative and postoperative complications, treatment and outcome were collected.

Results

The overall complication rate associated with metastatic iliosacral lymphadenectomy was 26.1%. The only intraoperative complication recorded was hemorrhage and was reported in 24 (17.6%) surgeries, 11 (45.8%) of which received a blood transfusion. Postoperative complications were reported in 10.4% of surgeries, and included edema formation (n = 4, 2.6%), unilateral or bilateral paraparesis (n = 4, 2.6%), hypotension (n = 3, 2.0%), surgical site infection (n = 2, 1.3%), abdominal incision dehiscence (n = 1, 0.6%), urinary incontinence (n = 1, 0.6%), and death (n = 1, 0.6%). The size of the iliosacral lymph nodes was significantly associated with a greater risk of complications, hemorrhage, and the need of transfusion during lymphadenectomy for metastatic apocrine gland anal sac adenocarcinoma.

Conclusion

Complications associated with iliosacral lymphadenectomy for metastatic apocrine gland anal sac adenocarcinoma are relatively common and mostly relate to hemorrhage. These complications are significantly associated with the size of the extirpated metastatic lymph nodes.

Clinical relevance

This retrospective study provides information for the clinician regarding the potential surgical complications for extirpation of metastatic iliosacral lymph nodes. These complications, although not common, can be severe and should be discussed with owners before surgery.

Introduction

Apocrine gland anal sac adenocarcinoma (AGASACA) is a locally aggressive and highly metastatic neoplasm (1). The regional lymph nodes (LNs), including medial iliac, internal iliac, and sacral lymph nodes, are often the earliest sites for metastasis. Metastatic disease has been reported in 36 to 96% of cases at the time of writing (1–7).

Surgery is considered the mainstay of treatment for dogs with AGASACA and locoregional disease. Lymph node extirpation in dogs with metastatic regional lymph nodes has been shown to prolong survival time (5,8).

The surgical complication rate for metastatic lymphadenectomy for AGASACA was reported as findings in 2 relatively small studies comprising a total of 79 cases. These 2 studies reported complications during and after lymphadenectomy to be 4.3% and 5%, respectively (9,10). Complications reported following iliosacral lymphadenectomy included severe hemorrhage, overflow urinary incontinence, celiotomy site dehiscence and infection, and death (1,9–11).

The iliosacral lymph center is composed of the medial iliac, internal iliac, and sacral LNs. The medial iliac lymph node, formerly known as the external iliac lymph node, is the largest of the iliosacral lymph center and it is consistantly located between the deep circumflex iliac and the external iliac arteries. It is usually a single LN on each side, but can be seen as a pair and only in one side of the body (12). This lymph node is deeply associated to the right side of the vena cava, which lies dorsal and to the right of the aorta. The internal iliac LNs, formerly called hypogastric LNs, are usually a pair of small nodes located at the angle between the internal iliac and median sacral artery. The sacral LNs are present in approximately 50% of dogs and when present, they lie ventral to the body of the sacrum on each side of the median sacral artery (12). Due to the close proximity to important vessels and location in the pelvic canal, iliosacral lymphadenectomy can be challenging. This may be especially important when removing large lymph nodes. In addition, important spinal nerves, such as femoral, obturator, and ischiatic nerves, exit the spinal cord in the ventral aspect of the vertebrae L6-S2 (12). Such nerves can run very close to the iliosacral lymph nodes when they are enlarged.

In these 2 studies, which report complications during surgical management for AGASACA, the authors determined that the only intra-operative complication for iliosacral lymphadenectomy was hemorrhage. The authors of the present study have experienced other complications in cases of large metastatic iliosacral LN extirpation. To the authors’ knowledge, the association between the size of the metastatic lymph nodes and intra- and postoperative complications for iliosacral lymphadenectomy has not been evaluated.

Reporting complications in veterinary soft tissue and oncologic surgery remains non-standardized. Follette et al (13) evaluated 151 articles and reported surgical complications and other adverse events in soft tissue and oncologic surgery in dogs and cats. Follette et al (13) determined that 92% of articles mentioned surgical complications, but only 7.3% defined the term complication. In the same study, terminology for defining adverse events, complications, and major reporting time frames was proposed, and the use of standardized grading schemes to report surgical complications was recommended.

The objective of this study was i) to report intra and immediate postoperative complications and short-term outcomes in dogs that underwent lymphadenectomy as part of cytoreductive surgery for AGASACA; and ii) to determine if the number and/or size of enlarged regional lymph nodes influence the occurrence of these complications.

Materials and methods

In this multi-institutional retrospective study, medical records of client-owned dogs that were diagnosed with AGASACA were reviewed. The inclusion criteria included i) patients that underwent iliosacral lymphadenectomy as part of cytoreductive surgery for AGASACA between January, 2010 and October, 2021; ii) histologically confirmed metastatic AGASACA (Stages 3a and 3b); iii) pre-operative staging including thoracic imaging, such as thoracic radiographs or computed tomography, abdominal imaging such as ultrasonography or computed tomography, and hematology (complete blood count and blood chemistry); and iv) at least 1-month postoperative follow-up.

Dogs were excluded if the extirpated lymph nodes were not confirmed to be metastatic or had neoplasia other than AGASACA upon histological examination and patients that were diagnosed with AGASACA but did not undergo iliosacral lymphadenectomy.

Clinical data retrieved included signalment, reproductive status, abdominal ultrasound or computed tomography findings, number of enlarged iliosacral lymph nodes, size of the largest iliosacral lymph node, pre-operative bloodwork abnormalities, surgical procedures performed, intra and postoperative complications, treatment for complications performed, histologic results, and short-term outcome (2-week survival).

All computed tomography (CT) scans and abdominal ultrasound (AUS) images were reviewed by Board-certified veterinary radiologists. Lymph node size measurements were taken from either CT scan, AUS, or histopathology report in that order of consideration when there were multiple measurements available. Maximum diameter of the largest iliosacral LN was recorded in both modalities (CT scan and AUS). This value was elected due to the variability on measurements available in CT or AUS reports, in which length, height, and width were reported inconsistently. The only measurement consistently reported was maximum diameter of the largest LN. Maximum diameter of tumor and metastatic LNs has been used to accurately predict prognosis in gastric carcinoma and colorectal carcinoma in human medicine (14,15).

Histopathology results confirming the presence of metastatic AGASACA in the lymph nodes for all cases were reviewed. These results were reported by Board-certified veterinary pathologists.

Surgical complications associated with iliosacral lymphadenectomy were retrieved from medical records. Surgical complications were defined as adverse events temporally associated with and attributed to surgical intervention (13). Complications were initially characterized as intra-operative and postoperative complications. As suggested by Follette et al (13), intra-operative complications were defined as those that occurred in the time from skin incision to skin closure and postoperative complications were defined as those that occurred after skin closure. Postoperative complications that occurred within the first 14 d were considered short term and the ones that occurred after 14 d were considered long term. Intra-operative complications were graded (I through IV) using the CLASSIC (classification of intra-operative complications) scheme for intra-operative complications (16). In this scheme, Grade 0 is defined as no deviation from the ideal operative course; Grade I includes any deviation from the ideal operative course, without the need for any additional treatment or intervention; Grade II includes any deviation from the ideal operative course, with the need for additional treatment or intervention, but that are not life-threatening and not leading to permanent disability; Grade III includes any deviation from the ideal operative course, with the need for additional treatment or intervention and is life-threatening and/or leading to permanent disability; and Grade IV includes any deviation from the ideal operative course causing the death of the animal.

Postoperative complications were classified according to the system used in human surgery; the Accordion Severity Classification of Postoperative Complications: contracted classification (17). According to this system, complications are classified in different “levels.” Level 1 or mild complications require only minor invasive procedures that can be performed at bedside, such as insertion of intravenous lines, urinary catheters, nasogastric tubes, and drainage of wound infections. Physiotherapy and the following drugs may be employed in Level 1; antiemetics, antipyretics, analgesics, diuretics, and electrolytes. Level 2 or moderate complications require pharmacologic treatment with drugs other than those used for minor complications (e.g., antibiotics); blood transfusions and total parenteral nutrition are also included. Level 3 or severe complications require endoscopic or interventional radiologic procedures or reoperation. These complications also include those that result in failure of one or more organ systems (17). Level 4 complications result in postoperative death.

Statistical analysis

Data were analyzed for normality using a Shapiro-Wilk test, which showed the data to be normally distributed. The difference between group means was evaluated using Student’s t-tests and a value of P ≤ 0.05 was considered significant for all analyses.

The cases were initially divided into 2 groups: surgical procedures that had complications (C) and surgical procedures that had no complications (NC). The cases were also divided into surgical procedures that had hemorrhage (H) and no hemorrhage (NH) and the ones that had hemorrhage were further divided into surgical procedures that had a blood transfusion (BT) and no blood transfusion (NBT).

The maximum diameters of the largest LNs (MDLLNs) extirpated in C group were compared to the NC group. The number of enlarged LNs (NELNs) extirpated in one surgery in C group were compared to the NC group. The MDLLN in the group of dogs that developed H was compared to the group with NH. The MDLLN in the group that developed H and had a BT was compared to the group that developed H and had NBT. The cases were also divided into Stage 3a and Stage 3b groups (metastatic LNs of < 4.5 cm and > 4.5 cm, respectively) (5) and compared to the cases that had complications (Group C).

Results

Cases (N = 151) were contributed by 5 institutions (University of Florida, The Ohio State University, University of Missouri, North Carolina State University, and BluePearl Pet Hospital Charlestown, Massachusetts, USA). One hundred and thirtysix dogs met the inclusion criteria and those case records were enrolled in the study. There were 43 breeds represented. Mixed-breed dogs (n = 40, 29.4%), Labrador retrievers (n = 16, 11.7%), and German shepherds (n = 7, 5.1%) were the most common. The mean age at the time of surgery was 9.4 y (range: 4 to 17 y). There were 91 (66.9%) castrated males, 43 (31.6%) spayed females, and 1 (0.7%) intact male. Hypercalcemia was seen pre-operatively in 43 dogs (31.6%), from which 14 dogs (10.3%) had only serum calcium measured, and 29 dogs (21.3%) had ionized calcium measured. The mean serum calcium value was 12.6 mg/dL (range: 10.5 to 17.4 mg/dL; reference: 8.7 to 10.4 mg/dL). The mean ionized calcium value of affected dogs was 1.64 mmol/L (range: 1.40 to > 2.5 mmol/L; reference: 1.25 to 1.45 mmol/L). Additional hematological abnormalities are summarized in Table 1.

Table 1.

Preoperative bloodwork abnormalities.

Bloodwork abnormality	Number of dogs	Percentage of dogs
Increased ALP	46	33.8%
Hyperproteinemia	34	25.0%
Increased ALT	32	23.5%
Neutrophilia	21	15.4%
Hyperglobulinemia	15	11.0%
Hypophosphatemia	13	9.5%
Increased BUN	13	9.5%
Increased AST	11	8.0%
Monocytosis	10	7.3%
Hypokalemia	8	5.9%
Hyperchloremia	7	5.1%
Anemia (low HCT)	6	4.4%
Increased Creatinine	6	4.4%
Neutropenia	5	3.7%
Hypochloremia	5	3.7%
Hyperkalemia	1	0.7%

The mean number of enlarged iliosacral LNs was 3 (median: 3, range: 1 to 7). The maximum diameter of the largest iliosacral LN was determined from medical records in 118 cases. The mean maximum diameter of the largest iliosacral LN was 4.7 cm (median: 4.0 cm; range: 0.7 to 17 cm).

All dogs had at least 1 procedure to remove metastatic iliosacral lymph nodes and 16 dogs had 2 or more sequential procedures to extirpate newly diagnosed metastatic LNs. A total of 153 surgical procedures were recorded.

A total of 40 (26.1%) complications associated with iliosacral lymphadenectomy are reported and summarized in Table 2. The only intra-operative complication recorded was hemorrhage, which was reported in 24 surgical procedures (N = 153, 15.7%). Twenty-four dogs developed hemorrhage, 11 (45.8%) of which received a blood transfusion.

Table 2.

Complications associated with iliosacral lymphadenectomy in 153 surgeries.

Intraoperative complications	n (%)
Grade II
Hemorrhage	13 (8.5)
Grade III
Hemorrhage	11 (7.2)
Postoperative complications
Level 1 — Mild
Edema formation (ventral or hind limb)	4 (2.6)
Paraparesis (unilateral or bilateral)	4 (2.6)
Urinary incontinence	1 (0.7)
Level 2 — Moderate
Hypotension	3 (2.0)
Surgical site infection	2 (1.3)
Level 3 — Severe
Incision dehiscence	1 (0.6)
Level 4 — Death
Euthanized 1-hour post-op	1 (0.6)

Sixteen (10.5%) postoperative complications were reported in 153 cases, which included edema formation in 4 dogs (2.6%), unilateral or bilateral paraparesis in 4 dogs (2.6%), hypotension in 3 dogs (2.0%), surgical site infection (SSI) in 2 dogs (1.3%), abdominal incision dehiscence in 1 dog (0.6%), urinary incontinence in 1 dog (0.6%), and death in 1 dog (0.6%). Three of the 4 cases of paraparesis resolved over the first 5 d after surgery and the single case of urinary incontinence resolved within the first 7 d after surgery.

Based on the Accordion Severity Classification of Postoperative Complications: Contracted Classification, postoperative complications were classified as Level 1 (n = 9, 56.2%); Level 2 (n = 5, 31.2%); Level 3 (n = 1, 6.3%); and Level 4 (n = 1, 6.3%). Based on the CLASSIC scheme, intra-operative complications associated with iliosacral lymphadenectomy included 13 (54.2%) Grade II hemorrhages and 11 (45.8%) Grade III hemorrhages.

One hundred and thirty-three dogs (97.8%) were discharged alive, and 3 dogs (0.02%) were euthanized within the first 7 d after surgery. One dog was euthanized during surgery because the primary anal sac tumor was invading a substantial portion of the rectum and the owner elected not to proceed with resection of the tumor; no complications were reported during iliosacral lymphadenectomy in this dog. The second dog became severely hypotensive and bradycardic on recovery, and was not responsive to 2 units of packed red blood cell transfusions and intravenous fluids. Humane euthanasia was elected. A third dog had postoperative arrhythmias, paraparesis, edema, incisional dehiscence, aspiration pneumonia, and hypotension refractory to vasopressors; the dog was euthanized 7 d after surgery while in hospital. There were no post-mortem examinations on any of these dogs.

In the analysis of MDLLN extirpated in C group compared to the NC group, the mean MDLLN was significantly larger in the C group (X̄ = 8.20 cm) compared with the NC group (X̄ = 3.83 cm) (P < 0.05). The number of NELNs removed in a single event was not significantly associated with greater complications or hemmorrhage (P > 0.05). The difference of the mean MDLLN between groups with hemorrhage with blood transfusion (X̄ = 10.63 cm) compared to hemorrhage with no blood transfusion (X̄ = 6.82 cm) was not statistically significant (P > 0.05). When the groups Stage 3a (LN < 4.5 cm) and 3b (LN > 4.5 cm) were compared, the risk of complications was significantly greater in LNs > 4.5 cm (P < 0.05).

Discussion

In the present study we determined that complications associated with iliosacral lymphadenectomy as part of cytoreductive therapy for AGASACA were relatively common. Intra-operative complications occurred in 15.7% of the surgical procedures. Postoperative complications occurred in 10.4% of the surgical procedures. These data differ from previous studies in which the rate of complications was lower. This could be related to the fact that even minor complications were included in the current study, in addition to our much larger cohort size. Furthermore, cohorts with both anal sacculectomy and iliosacral lymphadenectomy have been analyzed previous studies (9,10) and in those, there was a small number of lymphadenectomies. Despite complications being relatively common, most (56.2%) postoperative complications were considered mild; severe complications and death were only seen in 2 dogs. Regarding intra-operative complications, 54.2% were considered non-life-threatening and 45.8% were life threatening and required blood transfusions.

Hemorrhage was the only intra-operative complication which occurred during iliosacral lymphadenectomy. However, there were cases that had postoperative paraparesis or urinary incontinence that could have been caused by inadvertent damage to adjacent nerves, such as femoral, ischiatic, or pelvic nerves, during the surgical procedure. Similarly, hemorrhage has been reported to be the most common intra-operative complication in previous studies by Bennett et al (1), Sterman et al (9), and Barnes et al (10).

Hypercalcemia was discovered pre-operatively in 31.6% of the dogs herein. This finding is similar to what was seen in previous studies, in which hypercalcemia was reported in 25, 27, and 51% of dogs with AGASACA (1,3,18).

We detected that the maximum diameter of the iliosacral LNs was significantly associated with a greater risk of overall complications, hemorrhage, and the need for transfusion during lymphadenectomy for metastatic AGASACA. These findings are reasonable due to the close proximity of the iliosacral LNs to important vasculature and their location within the pelvic canal, which makes appropriate visualization and dissection challenging. In the cases that experienced paraparesis or incontinence, all but one dog recovered. These complications have not been previously reported. Furthermore, our study showed that cases in Stage 3b (LNs > 4.5 cm) had a higher risk of surgical complications compared with Stage 3a cases (LNs < 4.5cm). It is important to note that complications did occur when removing small LNs and these did not occur in most cases in which large LNs were being removed. Finally, our results showed that the number of LNs extirpated was not significantly associated with a greater rate of complications.

The overall mortality rate in our study was 2.2%. However, the cause for euthanasia could only be associated with the lymphadenectomy procedure in 1 case. To the authors’ knowledge, the mortality rate specific for iliosacral lymphadenectomy has not been reported in large studies. The only report of that nature is the small retrospective study by Bennett et al (1) in which it was determined that 1 of the 5 dogs that underwent iliosacral lymphadenectomy for metastatic AGASACA died from causes related to the procedure. In the authors’ experience the mortality rate for metastatic lymphadenectomy is very low. However, clients should be advised of the risks of hemorrhage prior to the extirpation of large iliosacral LNs. Previous studies showed an increase in survival times following lymph node metastatectomy for AGASACA (5,8). As an alternative, hypofractionated radiation therapy, alone or in combination with chemotherapy, has also been shown to lead to measurable and long-lasting responses (19,20). A prospective comparison between metastatic lymphadenectomy and hypofractionated radiation therapy has not been reported in the literature. In the retrospective study of Meier et al (19), the progression-free interval (PFI) and median survival times (MST) of dogs treated with radiation (3.8 Gy × 8) were significantly longer than those of dogs treated with surgery. The potential advantage offered by radiation therapy in that study should be viewed with caution. Interestingly, the author reported resolution of clinical signs in all dogs in the surgery group, whereas in the RT group, a partial response was seen in 5 dogs, stable disease in 2 dogs, and progressive disease in 2 dogs. Furthermore, regarding defecation-associated symptoms in the RT group, complete resolution was seen in 2 dogs, amelioration in 5 dogs, and stabilization in 1 dog (19). The longer PFI and MST in the radiation group despite the lack of confirmed complete responses is intriguing. Surgery and radiation have different treatment duration and costs. The potential advantages of surgery (single treatment, complete resolution of signs, and potentially cheaper) when compared to RT should be balanced with the fact that surgery seems to carry a greater risk of major complications and may provide a shorter survival. We believe in the advantages and disadvantages of surgery and RT should be discussed with clients allowing for the creation of a plan that best fits each patient as well as the clients’ capabilities.

The use of more readily available advanced imaging, such as CT scans may lead to better outcomes and lower complication rates due to better presurgical planning. The use of these diagnostic modalities is recommended for the evaluation of metastatic AGASACA.

The primary limitations of our study are related to its retrospective nature, which include the lack of standardization of documentation of complications and choice of diagnostic imaging (CT scan versus AUS), which leads to non-standardized LNs measurements. Since CT has a higher sensitivity for detecting iliosacral LNs, it is possible that the number of enlarged LNs in dogs that had AUS was underestimated (21,22). In addition, CT scans and AUS were evaluated by different Board-certified radiologists, which can lead to examiner-related biases.

In conclusion, complications associated with iliosacral lymphadenectomy for metastatic AGASACA were relatively common but most of them were considered mild, and the most common complication was hemorrhage. Severe complications can occur and seem to be more common when extirpating metastatic LNs > 4.5 cm. This retrospective study provides information for the clinician regarding the potential surgical complications for extirpation of metastatic iliosacral LNs. CVJ