Partial splenectomy for incidentally detected non-ruptured splenic lesions in dogs: 18 cases (2004–2018)

Abstract

The objective of this retrospective study was to describe the outcome and incidence of splenic malignancy in 18 dogs undergoing partial splenectomy for incidentally detected, non-ruptured splenic lesions. Incidence of splenic malignancy in the present study was 5.6% [95% confidence interval (CI): 0.14% to 27.65%]. Median diameter of splenic nodules was 2 cm (range: 1.5 to 4 cm). Splenic hemangiosarcoma was diagnosed in 1 dog, while the remaining 17 dogs had benign splenic lesions. There was a higher incidence of non-splenic malignancy (50%) than splenic malignancy (5.6%) in the study population. Overall median survival time after surgery was 300 days (range: 4 to 1332 days). Median survival time in dogs with malignant disease (splenic and non-splenic) was 67 days (range: 4 to 425 days) and for non-malignant disease was 727 days (range: 8 to 1332 days). In conclusion, partial splenectomy may be appropriate for small, incidental non-ruptured splenic lesions in dogs.

Introduction

Splenic lesions are commonly encountered in canine patients. Many of these lesions are incidentally detected on imaging or at surgery. Various methods of differentiating between malignant and benign splenic lesions have been studied to predict survival time in dogs with splenic disease. These methods include needle aspirate cytology, advanced imaging techniques, microRNA expression profiles (1–4), and potential prognostic indicators such as hemoabdomen, anemia, and mass to body weight ratio (5–10). Most localized splenic lesions are benign (11,12); however, the incidence of malignancy in dogs with splenic masses is as high as 59% (5,6,13). Higher rates of malignancy, especially hemangiosarcoma, are encountered in cases with splenic rupture and associated hemoperitoneum (7,9,10). The difficulty in distinguishing malignant from nonmalignant lesions and risk assessment for organ removal versus preservation, generally sways the decision toward performing a total splenectomy rather than a partial splenectomy. Partial splenectomy preserves splenic function and has been reserved for cases with focal non-malignant diseases such as abscesses and splenic trauma, whereas its use in cases of malignant splenic neoplasia is considered inappropriate (14).

Although indications, contraindications, and different techniques for performing partial splenectomy have been reported in the dog (14–18), to the authors’ knowledge, there are no reports describing the clinical application and outcome of this procedure. The objective of this retrospective study was to report the incidence of splenic malignancy and survival times for dogs that underwent partial splenectomy for incidentally detected, non-ruptured splenic lesions. Based on the speculation that low clinical suspicion for splenic malignancy and inherent bias may have influenced case selection, we hypothesized that frequency of splenic malignancy in dogs undergoing partial splenectomy for incidentally detected non-ruptured splenic lesions would be < 30% as previously reported (12). We further hypothesized that median life expectancy in this study population would be comparable to the median life expectancy previously reported for dogs undergoing total splenectomy for incidentally detected non-ruptured splenic nodules (12).

Materials and methods

Electronic medical records from the small animal veterinary teaching hospitals at Auburn University and Texas A & M University were searched to identify dogs that underwent partial splenectomy for incidentally detected splenic lesions between January 1, 2004 and June 30, 2018. The criteria for inclusion were complete medical records including a surgery report, histopathology report, and documentation that there was no evidence of splenic rupture or hemoperitoneum attributed to splenic hemorrhage.

Data regarding age, breed, gender, body weight, and presenting clinical signs were obtained for each patient. Results of abdominocentesis, thoracic and abdominal radiographs, abdominal ultrasonography, computed tomography, and cytologic diagnosis of ultrasound-guided needle aspirates were included when available. Abdominal ultrasonography and interpretation of imaging studies were performed by Board-certified radiologists or by radiology residents and confirmed by supervising Board-certified radiologists.

Data related to surgical findings, procedures performed, tissues biopsied, estimated number of splenic nodules and largest diameter of splenic nodules were collected from surgical or ultrasonographic reports. All surgeries were performed by American College of Veterinary Surgeons (ACVS) Board-certified surgeons or surgical residents under direct supervision of ACVS Board-certified surgeons.

All cytologic and histologic examinations were performed by Board-certified pathologists or by pathology residents and confirmed by supervising Board-certified pathologists.

Survival data were obtained by reviewing records for euthanasia dates or last date of patient examination or recheck at the study facilities. When applicable, primary care veterinarians were contacted to obtain date of euthanasia or spontaneous death or the date of the last medication refill or client follow-up. Additional data regarding the cause of death, diagnosis at the time of euthanasia, and whether the patients with malignant disease underwent chemotherapy or additional surgery were not recorded or available.

Data were summarized as median and range or number and percentage for continuous or categorical parameters, respectively. Incidence of splenic malignancy was reported in percentage and 95% confidence intervals (CI) were calculated using the modified Wald method (19).

Results

The electronic medical record search identified 19 dogs that underwent partial splenectomy for incidentally detected splenic nodules during the study period. One dog was excluded due to a missing histopathology report. Eighteen dogs met the inclusion criteria including 1 dog that had hemoperitoneum secondary to hepatic mass rupture. Of the 18 dogs, 9 were spayed females, 7 were castrated males, and 1 each was an intact female and an intact male. Mixed breed dogs were most commonly represented (4/18) followed by bulldogs (2/18), and 1 each of various other breeds. Median age and body weight at the time of surgery were 12 y (range: 8 to 15 y) and 13.6 kg (range: 6.1 to 37.4 kg), respectively; with 66% of dogs weighing less than 20 kg.

Most of the patients showed nonspecific clinical signs such as lethargy (n = 15), anorexia (n = 10), vomiting (n = 4), abdominal pain (n = 3), abdominal distention (n = 3), polyuria/polydipsia (n = 2), ataxia (n = 1), and weight loss (n = 1). Thoracic radiographs, abdominal radiographs, and abdominal ultrasound were performed in 14, 9, and 13 dogs, respectively. Most dogs were evaluated using more than 1 imaging modality. Computed tomography (CT) was pursued in 3 dogs with 1 dog each receiving thoracic and abdominal CT, head and abdominal CT, and abdominal CT, respectively. Incidental splenic nodules were identified in 11/13 dogs (85%) by abdominal ultrasound; 1 dog each had generalized splenomegaly and a spleen that was deemed normal on ultrasound. Ultrasound-guided splenic aspirates were performed in 4/13 dogs; 2 dogs had identical cytologic and histopathologic diagnosis of lymphoid hyperplasia and in the other 2 dogs cytologic diagnosis of lymphoid hyperplasia was made while the histopathologic diagnosis revealed a splenic hematoma. Abdominocentesis with fluid analysis was performed in 4 dogs, identifying 3 dogs with septic abdomen and 1 with bile peritonitis. Incidental splenic nodules were detected on pre-operative imaging in 14/18 dogs; 4 dogs had splenic nodules detected at exploratory celiotomy.

All dogs in this series had an independent, non-splenic indication for exploratory celiotomy. Primary indications for celiotomy included diagnosis of a non-splenic abdominal mass in 10/18 dogs, septic peritonitis in 3 dogs, bile peritonitis in 2 dogs, intestinal obstruction secondary to a foreign body in 1 dog, colopexy for perineal hernia in 1 dog, and gastric dilatation and volvulus in 1 dog. Accordingly, all 18 dogs had a partial splenectomy performed and 1 or more procedures performed for non-splenic pathology (Table 1). A solitary splenic nodule was identified in 16/18 dogs, while 1 dog had 2 similarly sized nodules, and 1 dog had 3 similarly sized nodules. Size of splenic nodules was available in 15/18 dogs, and the median diameter of splenic nodules was 2 cm (range: 1.5 to 4 cm). Partial splenectomy was performed using a crush-suture technique (7 dogs), thoraco-abdominal stapler (TA 55 = 3 dogs, TA 30 V3 = 3 dogs), or a vessel sealing device (LigaSure = 1 dog). The partial splenectomy technique was not specified in 4 dogs. No intra-operative or post-operative complications attributed to partial splenectomy were documented in the medical record.

Table 1.

Histopathologic findings and survival times in dogs undergoing partial splenectomy and concurrent procedures.

Dogs	Signalment	Partial splenectomy + other procedures performed	Histopathologic findings — spleen	Histopathologic findings − other organs	Survival (days)
1	10 y MC mixed breed	Colopexy	NLH, EMH	None	727
2	9 y FS basset hound	Gastropexy	Congestion, hematoma	None	1307
3	13 y FS mixed breed	Liver lobectomy, adrenalectomy, liver biopsy	Infarction	Adrenal cortical carcinoma,22 hepatocellular carcinoma
4	14 y FS basenji	Ovariohysterectomy	NLH, infarction	Ovarian adenocarcinoma	NA
5	12 y FS corgi	Bilateral adrenalectomy, liver biopsy	Hematoma	Bilateral adrenocortical adenoma peliosis hepatis	NA
6	13 y FS Irish terrier	Pancreatic mass removal, liver biopsy, J-tube placement	NLH	Insulinoma, metastatic insulinoma	33
7	13 y MC springer spaniel	Hepatic mass removal, liver biopsy	Low grade HSA	Hepatic abscess	413
8	10 y FS Boston terrier	Pancreatic mass removal, adjacent lymph node	Subcapsular haemorrhage, HSD	Insulinoma, metastatic insulinoma	425
9	15 y MC Jack Russell terrier	Gastrotomy	NLH	None	612
10	11 y FI miniature poodle	Hysterectomy, jejunal resection anastomosis, omentectomy	Capsular metastatic intestinal adenocarcinoma, EMH	Jejunal mucinous adenocarcinoma, carcinomatosis, metastatic adenocarcinoma	4
11	10 y FS bulldog	Jejunal resection anastomosis	NLH, EMH	Undifferentiated sarcoma-jejunal mass	80
12	12 y FS bulldog	Cholecystectomy, liver biopsy	Myelolipoma with intra-lesional, HSD, EMH	Gall bladder infarction, Cholangitis	269
13	13 y MI mixed breed	Partial liver lobectomy	Splenic abscess, NLH, EMH	Hepatic abscess	8
14	10 y MC West Highland white terrier	Jejunal resection anastomosis	Infarction and EMH, fibrino-necrotizing capsulitis	None	785
15	8 y MC golden retriever	Liver lobectomy, gastropexy	NLH	Soft tissue sarcoma-liver, peritoneal mass	67
16	10 y MC dachshund	Partial liver lobectomy	Congestion neutrophilic serositis steatitis	Hepatic abscess	1332
17	12 y mixed breed	Liver lobectomy, adrenalectomy	Hematoma	Hepatocellular carcinoma, adrenal cortical carcinoma	330
18	13 y FS beagle	Liver lobectomy	Infarction	Hepatocellular carcinoma	28
Median					300

MC — male castrated; NLH — nodular lymphoid hyperplasia; EMH — extramedullary hematopoiesis; FS — female spayed; NA — Not available; HSA — hemangiosarcoma; HSD — hemosiderosis; FI — female intact; MI — male intact. congestion

On histopathology, 10/18 dogs had evidence of malignant neoplasia in at least 1 abdominal organ (Table 1). In 4 dogs spleen was the only tissue submitted; whereas in 14 dogs other tissue(s) were also submitted. Two concurrent primary malignant non-splenic neoplasms, hepatocellular carcinoma, and adrenocortical carcinoma were detected in 2 dogs. Solitary primary malignant non-splenic neoplasia was diagnosed in 7 dogs, including insulinoma in 2 dogs, hepatocellular carcinoma in 1 dog, jejunal mucinous adenocarcinoma in 1 dog, jejunal undifferentiated sarcoma in 1 dog, papillary and cystic adenocarcinoma of the ovary in 1 dog and high grade sarcoma of undetermined cell of origin in the liver in 1 dog. Metastatic disease was diagnosed in 4 dogs based on histopathological evaluation of liver biopsies, abdominal lymph nodes, omentum, or peritoneum.

In 5/18 dogs with non-splenic pathology in which no evidence of malignancy was found, 3 dogs had hepatic abscess, 1 dog had gall bladder wall infarction, and 1 dog had bilateral adrenocortical adenoma.

The incidence of splenic malignancy was 5.6% (95% CI: 0.14% to 27.65%). Splenic hemangiosarcoma was diagnosed in 1 dog, while the remaining 17 dogs had benign lesions. In the dog with adenocarcinoma of the ovary, histopathology of the spleen revealed 1 microscopic focal perivascular accumulation of spindle-shaped cells of uncertain clinical significance. Additionally, in the dog with intestinal mucinous adenocarcinoma with gross abdominal carcinomatosis, neoplastic cell invasion of the splenic capsule was noted.

Benign splenic lesions in this study population comprised solely or a combination of lymphoid nodular hyperplasia, congestion and hematoma, infarction, extramedullary hematopoiesis, myelolipoma, hemosiderosis, capsulitis, serositis, abscess, and infarction (Table 1).

Overall median survival time after surgery was 300 d (range: 4 to 1332 d). Median survival time in dogs with non-malignant disease was 727 d (range: 8 to 1332 d), whereas that for dogs with malignant disease of splenic and non-splenic tissues was 67 d (range: 4 to 425 d). Amongst dogs with malignant disease, 1 dog had hemangiosarcoma with 357 d between partial splenectomy and last follow-up, while the rest of the dogs had non-splenic malignancy.

Discussion

To our knowledge this is the first study to describe outcome following partial splenectomy for incidentally detected, non-ruptured splenic nodules in dogs. In the present study, dogs with incidentally detected, non-ruptured splenic nodules had a higher frequency of benign lesions (17/18, 94%) than malignant tumors (1/18, 6%). Hemangiosarcoma was the only malignant splenic neoplasm diagnosed in 1 of 18 dogs (6%); this frequency is lower than the 30% incidence of splenic malignancy previously reported for cases with incidentally detected, non-ruptured splenic nodules (12). The lower incidence of splenic malignancy in the present study may be attributed to selection bias, small patient population and lower median body weight of the study population. In the present study 66% of dogs weighed < 20 kg; dogs weighing ≤ 28 kg are reportedly less likely to be diagnosed with splenic hemangiosarcoma (20). In 1 study, larger splenic masses were more likely to be benign, while smaller masses were more likely to be malignant (7). However, in the present study the median diameter of the splenic lesions was 2 cm (range: 1.5 cm to 4 cm), with all but 1 splenic lesion being benign. There was a higher incidence of non-splenic malignancy (50%) than splenic malignancy (5.6%) in the present study population.

Surgery is the primary method of treatment of incidentally detected non-ruptured splenic lesions and allows accurate histopathologic diagnosis, resection of macroscopic malignant disease, cure for benign lesions, and may forestall the risk of splenic rupture. Clinical suspicion for splenic malignant neoplasia when non-ruptured splenic nodules are incidentally detected and consequently the decision to perform partial splenectomy are based on surgeon’s assessment of signalment, history, presentation, clinical signs, associated co-morbidities, imaging findings, cytologic findings, intra-operative findings and gross appearance, anesthesia concerns, total operative/anesthesia times and inherent bias. Waldron and Robertson (16) reported that partial splenectomy can be performed quickly, less than 1 min with a thoraco-abdominal stapler, and approximately 5 min by suturing technique. We speculate that a low clinical suspicion for splenic malignant neoplasia allowed for consideration of partial splenectomy with a goal of minimizing total operative and anesthetic times in this study population. However, records of total surgical times were not available, and this is a limitation of the study.

The 56% overall incidence of malignancy in splenic and non-splenic tissues seemingly influenced the median survival time of 300 d in the present study. However, median survival time of dogs with non-malignant disease in the present study population was 727 d, being longer than the 436-day survival time previously reported for dogs with incidentally detected, non-ruptured benign splenic nodules (12). The difference in median survival time may be attributed to the small patient population in the present study. Since a single case of splenic hemangiosarcoma was diagnosed in the present study, influence of partial splenectomy on survival in cases with splenic malignant neoplasia could not be evaluated.

The present study is a bi-institutional retrospective study in which no predetermined case selection criteria were identified; however, noteworthy characteristics of the study population include median body weight of 13.6 kg with 66% of dogs weighing < 20 kg, predominantly solitary nodules with a median diameter of 2 cm, incidental detection, and no evidence of splenic rupture. The retrospective nature of the study, absence of a total splenectomy-control group, small patient population, high incidence of non-splenic malignancy, and inadequate data on survival in cases with splenic malignancy that underwent partial splenectomy do not allow any case selection criteria to be proposed and may warrant further research.

Total splenectomy is the preferred surgical treatment in cases in which malignant splenic neoplasm is suspected. The results of our study suggest that partial splenectomy may be appropriate for small, incidentally detected non-ruptured splenic masses. However, the potential benefits and risks should be assessed independently for every case.

The limitations of the present study include small sample size, those inherent to retrospective studies, lack of total operative times, and method of survival data collection. We were not able to investigate the influence of confounding variables such as concurrent malignant processes in non-splenic abdominal organs on the decision to perform a partial splenectomy. Additionally, a patient population without any non-splenic disease would have allowed a better evaluation of survival time following partial splenectomy for incidentally detected splenic lesions.

In conclusion, partial splenectomy may be appropriate for small incidentally detected non-ruptured splenic lesions in dogs.