Abstract
Two cases of secondary, inappropriate polycythaemia caused by renal adenocarcinoma in domestic shorthair cats, are described. The cats were 9 and 12 years old and both were presented because of generalised seizures presumably due to hyperviscosity. Both cats had a markedly increased haematocrit (0.770 and 0.632 l/l) and thrombocytosis (744×109/l and 926×109/l). An abdominal ultrasound revealed a mass in the cranial pole of one kidney in both cats. Serum erythropoietin (EPO) concentration was within the reference interval (RI) in both cats but was inappropriately high considering the markedly increased haematocrit. The cats were initially stabilised and managed by multiple phlebotomies and intravenous fluid therapy and underwent nephrectomy of the affected kidney later on. Both the polycythaemia and thrombocytosis resolved following surgery. Postoperative serum EPO concentration, measured in one cat, decreased markedly. Histopathology of the affected kidneys confirmed a diagnosis of renal adenocarcinoma. Both cats were stable for an 8-month follow-up period; however, one cat had developed a stable chronic kidney disease (CKD), while the other was represented 8 months postoperatively due to dyspnoea, and had radiographic evidence of lung metastasis, presumably because of the spread of the original renal tumour and was euthanased. Initial stabilisation of polycythaemic cats should include multiple phlebotomies. Nephrectomy should be considered in cats with secondary, inappropriate, renal adenocarcinoma-related polycythaemia when only one kidney is affected by the tumour, and provided that the other kidney's function is satisfactory. Nephrectomy should be expected to resolve the polycythaemia and lead to normalisation of serum EPO concentration.
Polycythaemia is defined as increased red blood cell (RBC) count above reference interval (RI), and is classified as relative or absolute (Erslev 1995, Messinezy and Pearson 1999, McMullin et al 2005). Relative (apparent) polycythaemia is characterised by increased packed cell volume (PCV) and RBC concentration, normal body RBC mass and decreased plasma volume, and most commonly results from dehydration or third body space losses. In absolute polycythaemia the PCV, RBC count and total body RBC mass are increased. It is further classified into primary and secondary polycythaemias. Primary polycythaemia, or polycythaemia (rubra) vera, is a myeloproliferative disorder previously described in cats, and is erythropoietin (EPO)-independent (Foster and Lothrop 1988, Watson et al 1994, Evans and Caylor 1995). In secondary absolute polycythaemia, the increase in RBC mass develops in response to EPO or EPO-like substances, and can be appropriate or inappropriate (Giger 1992, Erslev 1995, McMullin and Percy 1999, Donnelly 2001). In the appropriate form, increased plasma EPO concentration results from generalised tissue hypoxia, which is absent in the inappropriate form (Erslev 1995).
Most cases of feline secondary polycythaemia are appropriate, mainly as a result of congenital heart defects with right-to-left blood shunting, chronic pulmonary diseases and hyperthyroidism and less commonly from arterial hypoxia of other causes, neoplasia and renal disease (Kirby and Gillick 1974, Watson et al 1994, Evans and Caylor 1995, Stockham and Scott 2002, Yuki et al 2005).
To the best of our knowledge, only five cases of feline primary renal tumour-related inappropriate polycythaemia have been previously reported, three of which were only briefly mentioned in two large-scale surveys of feline renal neoplasia and serum EPO concentrations, respectively (Hasler and Giger 1996, Henry et al 1999). Two cases of renal adenocarcinoma-associated polycythaemia were previously reported, however, only the abstract is available in English (Yuki et al 2005). In the present report, we describe two cases of feline primary renal adenocarcinoma-related secondary inappropriate polycythaemia and their resolution following nephrectomy of the affected kidney.
Case 1, a 9-year-old, intact male, domestic shorthair cat was referred to the Hebrew University Veterinary Teaching Hospital (HUVTH) with a chief complaint of several generalised seizure episodes during the previous 3 months. Upon presentation, it had a 7/9 body condition score, was severely lethargic and depressed, with hyperaemia of the skin, mucus membranes and conjunctivae. Complete blood count (CBC) revealed marked polycythaemia (RBC 20.16×1012/l), increased haematocrit (0.77 l/l), haemoglobin concentration (223 g/l, RI 80–150) and thrombocytosis (744×109/l, RI 160–700×109/l). Serum biochemistry revealed hyperglycaemia (14.49 mmol/l, RI 3.89–6.11), hyperproteinaemia (total protein (TP) 84 g/l, RI 55–75), increased urea concentration (9.83 mmol/l, RI 3.49–6.98), and increased serum activities of aspartate aminotransferase (AST, 116 U/l, RI 14–50) and creatine kinase (CK, 5592 U/l, RI 13–100). Creatinine concentration was within RI (104.31 μmol/l, RI 35.36–132.60). The systolic arterial blood pressure was 150 mmHg. Urinalysis revealed haematuria (50 RBC/μl), proteinuria (0.5 g/l), and specific gravity (USG) of 1.020. Serum EPO concentration was 12.35 mU/ml (RI 10–30 mU/ml).
Abdominal ultrasound revealed a 2.5×2.5 cm hypoechoic, round mass, within the cranial pole of the right kidney, but was otherwise unremarkable. Thoracic radiographs were unremarkable.
The cat was hospitalised and treated with intravenous lactated Ringer's solution (LRS) and multiple phlebotomies (10–15 ml q6–8 h for 2 days). The PCV and the TP gradually declined to 0.6 l/l and 60 g/l, respectively. On day 3, an exploratory laparotomy was performed as the renal mass was assessed to be the cause of polycythaemia. The only abnormality noted on laparotomy was a round 3-cm intra-renal mass in the cranial pole of the right kidney. Nephrectomy was performed, and the kidney was submitted for histopathology.
Over the postoperative hospitalisation period (4 days), a gradual decline in the PCV was observed (0.44 l/l on day 4 postoperatively). Although the cat was mildly azotaemic (creatinine 194.48 μmol/l), it was discharged at the owner's request. It was rechecked 9 days later. The PCV, TP and platelet count were 0.35 l/l, 78 g/l and 363×109/l, respectively, however, the cat, although mentally improved, was still depressed, anorexic and azotaemic (creatinine 216.58μmol/l), and was thus hospitalised for 4 additional days and treated with intravenous LRS. It had improved and was discharged with a PCV of 0.39 l/l and creatinine concentration of 190.06 μmol/l. Twenty-six days later, it was rechecked and found to be mentally normal, hydrated, and had regained its appetite, however, owners reported polyuria and polydipsia. The PCV was 0.24 l/l, and creatinine concentration was 309.40 μmol/l and thus the cat was diagnosed with chronic kidney disease (CKD). Eight months later, the cat presented a 6/9 body condition score, with a stable stage III CKD (creatinine concentration 265.20 μmol/l), based on the International Renal Interest Society classification scheme (Polzin et al 2005), a PCV of 0.25 l/l and a TP of 70 g/l.
Case 2, a 13-year-old, female spayed, domestic shorthair cat that lived entirely indoors, was presented to the HUVTH with a chief complaint of a cluster of generalised tonic–clonic seizures, first noticed by the owners a few hours before presentation, and was treated by the referring veterinarian with diazepam (Assival; Teva, 1 mg/kg IV).
CBC revealed marked polycythaemia (RBC 17.9×1012/l) and increased haematocrit (0.632 l/l), haemoglobin concentration (221 g/l) and thrombocytosis (926×109/l), while the only serum biochemistry abnormalities were increased serum activities of AST (81 U/l) and CK (2544 U/l). Venous blood gas analysis was within RI. Systolic blood pressure was 140 mmHg. Serum EPO concentration was 29.43 mU/ml.
Abdominal ultrasound revealed a 3×3 cm, round, smooth, hypoechoic mass, bulging from the cranial pole of the left kidney. Thoracic radiographs were unremarkable. As polycythaemia was presumed to be the cause of seizures, multiple phlebotomies were performed during the next 48 h, and a total volume of 100 ml of blood was replaced with intravenous LRS, and by day 2 the PCV decreased to 0.52 l/l.
Exploratory laparotomy was performed a week later and revealed no abnormalities with exception of a well-defined encapsulated mass on the cranial pole of the left kidney. Nephrectomy was performed, and the kidney was submitted for histopathology. The cat recovered uneventfully and was discharged a day later with a PCV of 0.42 l/l.
Twenty-five days postoperatively, the cat was reported to be normal. Physical examination was unremarkable, and CBC as well as serum biochemistry analysis was within RI (PCV of 0.44 l/l, TP 77 g/l and platelets 410×109/l). Serum EPO concentration had decreased to 5.05 mU/ml.
The cat was represented 8 months later, with chief complaints of anorexia, weight loss and tachypnoea lasting for 2 weeks. Physical examination revealed tachypnoea with muffled respiratory and heart sounds. Thoracic radiography revealed a diffuse pulmonary nodular pattern, and mild pleural effusion. Metastatic lung disease (MLD) secondary to the renal adenocarcinoma was suspected, and the cat was euthanased at the owner's request. Due to the owner's objection, a post-mortem examination could not be performed.
Histopathology of both affected kidneys revealed large, densely cellular, well-demarcated, nodular expansile, non-capsulated processes of roughly 2×2.5×3 cm size in the cranial renal cortical poles. The tissue of both masses consisted of tubular structures with very fine basal membrane and well-differentiated epithelial cells resembling well-differentiated tubular epithelium and a lumen, presenting slightly larger than normal round nuclei each containing 1–2 small nucleoli and finely granular eosinophilic cytoplasms producing abundant luminal amphophilic secretum with rare mitotic figures (Fig 1). Between the tubular structures, interstitial blood vessels were present. Remnants of the original, pre-existing normal renal tissue were observed within the mass in cat 2 with occasional trapped normal-looking glomeruli among the neoplastic cells (Fig 2). The morphological diagnosis in both samples was primary renal adenocarcinoma.
Cat 1: microscopic cross-section through the renal mass. Well-differentiated neoplastic cells are arranged in a tubular pattern. The normal renal tissue is compressed by the tumour (arrow). The morphological diagnosis was renal adenocarcinoma (haematoxylin and eosin).
Cat 2: microscopic cross-section through the renal mass. The neoplastic cells are arranged in multiple tubular-like structures. Several, normal-looking glomeruli of the remaining normal renal parenchyma are noticed, trapped within the neoplastic tissue (arrow). The morphological diagnosis was renal adenocarcinoma (haematoxylin and eosin).
Primary renal neoplasia, as well as secondary absolute polycythaemia, is rare in cats, and the latter has been reported occasionally (Kirby and Gillick 1974, Crow 1985, Cook and Lothrop 1994, Klausner and Caywood 1995, Henry et al 1999, Meuten 2002, Yuki et al 2005). Renal neoplasia has been associated with secondary polycythaemia in people and occasionally in dogs (Peterson and Zanjani 1981, Nelson et al 1983, Gorse 1988, Waters and Prueter 1988, Da Silva et al 1990, Crow et al 1995, Erslev 1995, Bennett 2004), but was only briefly discussed in cats (Hasler and Giger 1996, Henry et al 1999, Yuki et al 2005). In both present cats, the association of the polycythaemia with renal neoplasia was based on exclusion of cardiopulmonary causes of general tissue hypoxia, due to lack of dyspnoea or cyanosis, normal thoracic radiography, a histopathological diagnosis of renal adenocarcinoma, and resolution of polycythaemia after nephrectomy of the affected kidney.
The chief complaint in both polycythaemic cats presented here included generalised seizures that in case 1 had recurred several times over 3 months. Neurological abnormalities, such as seizures, were previously reported in feline polycythaemia (Watson et al 1994, Evans and Caylor 1995), presumably due to hyperviscosity, although blood viscosity determination was unavailable.
In addition to haematological evidence of polycythaemia, thrombocytosis was present in both cats upon presentation. Thrombocytosis has never been reported in feline secondary polycythaemia. It might have resulted from an inflammatory response (eg, reactive), paraneoplastic process, or EPO-induced megakaryopoiesis (Burstein and Breton-Gorius 1995, Stockham and Scott 2002). Normalisation of the platelet count, in both cats, after nephrectomy suggests that this thrombocytosis was associated with renal neoplasia, however, its pathogenesis is unclear.
Cat 1 probably had a stage I CKD (regardless of the mass) at the time of presentation, as the creatinine concentration upon admission was 104.31 μmol/l, with USG of 1.020 in face of dehydration. The surgery, anaesthesia and the postoperative recovery period might have induced further renal injury as serum creatinine did increase to 309.40 μmol/l 26 days later. Moreover, serum creatinine did not decrease over the following 8 months, suggesting that hypertrophy of the other kidney did not occur, presumably due to an underlying CKD. Further preoperative diagnostics, including glomerular filtration rate (GFR) measurement, with assessment of the relative contribution of each kidney to the overall GFR (ie, scintigraphy), might have revealed a pre-existing CKD, and might have provided an assessment of the risk of developing azotaemia following a unilateral nephrectomy; thus should be considered in such cases.
In both present cats EPO concentrations were within RI, although in cat 2 it was at the upper RI. Low to normal serum EPO concentrations are characteristic of primary polycythaemia (Giger 1992, Hasler and Giger 1996), while in secondary polycythaemia its concentrations are expected to be increased (Giger 1992, Hasler and Giger 1996, McMullin and Percy 1999, Donnelly 2001, Yuki et al 2005). Although a previous study of feline renal adenocarcinoma-related secondary inappropriate polycythaemia has reported increased serum EPO concentration (Yuki et al 2005), its normal concentrations in both of our cats are not surprising, and are in agreement with several studies of feline polycythaemia (Cook and Lothrop 1994, Hasler and Giger 1996). However, in both present cats, these were inappropriately high in the face of an ongoing polycythaemia. These apparently normal serum EPO concentrations possibly resulted from cyclic EPO secretion, production of an EPO-like substance unrecognisable in the currently used enzyme-linked immunosorbent assay (ELISA) (although such substance has yet to be described) and decreased EPO production due to an existing partially normal feedback mechanism with renal oxygenation (Evans and Caylor 1995, Hasler and Giger 1996). Adaptive renal mechanisms to hypoxia rather than feedback inhibition of EPO may also explain the normal serum EPO values, as has been suggested in chronically hypoxaemic human patients (Eckardt et al 1990).
The source of EPO in secondary renal neoplasia-associated polycythaemia remains unclear. Two theories have been suggested. One suggests that renal tumour cells inappropriately produce and secrete excess EPO. Alternatively, the tumour may induce renal hypoxia via compression of the normal renal parenchyma and renal blood flow impairment, thereby inducing normal renal EPO-producing cells to amplify their EPO production (Waters and Prueter 1988, Bennett 2004).
Feline primary renal neoplasia has been reported to be aggressive and have a high metastatic rate, and is thus considered to have a grave prognosis (Crow 1985, Klausner and Caywood 1995, Henry et al 1999, Meuten 2002). The prognosis of secondary polycythaemia depends on the ability to control the primary disease. In the present cats, the polycythaemia has resolved after nephrectomy, and the clinical abnormalities (eg, seizures) never reappeared. Thus, despite the above-mentioned grave prognosis of feline renal neoplasia, nephrectomy seems to be warranted and should be considered in such cases, providing that MLD is absent (Yuki et al 2005). An underlying CKD, as was evident in cat 1 following unilateral nephrectomy, should be a concern. Careful, sequential phlebotomies and supportive therapy, as provided in these cases, were a good temporary preoperative solution for the management of polycythaemia.