Abstract
A 3-year-old cat was presented with increasing dyspnoea over the past four days. Unilateral pleural effusion was diagnosed and a modified transudate was drained several times. Surgical exploration revealed intra-thoracic prolapse of the left kidney and partial herniation of the spleen through a dorsal, circumferential diaphragmatic tear. Biochemical analysis of the pleural fluid confirmed urothorax. Due to excessive fibrin deposit on the well-vascularised kidney it was impossible to re-establish left urinary pathways. Left-sided nephrectomy and diaphragmatic herniorrhaphy were performed. Postoperative recovery was uneventful and complete. This is the first report of an urothorax in veterinary medical literature.
A 3-year-old, 3.3 kg, spayed female, domestic shorthair cat was referred for evaluation of increasing dyspnoea over the past four days. The cat had experienced slight dyspnoea several months previously, but seemed normal the weeks preceding this episode. The cat was allowed outside and was currently vaccinated (feline leukaemia, panleucopenia, cat flu). The referring veterinarian had taken a lateral radiograph showing an extensive fluid or soft tissue obliteration in the thoracic cavity. The partly radiographed abdominal cavity showed no significant abnormalities, with the exception that only one renal shadow was visible (Fig 1). Prior to referral, the cat received no medication.
00097-9/unzip/10.1016_S1098-612X(02)00097-9-fig1.png)
Lateral radiograph of a cat. Note the extensive fluid or soft tissue obliteration of the ventral thoracic cavity and the single renal silhouette visible.
On physical examination, the cat was lethargic and dull, and showed inspiratory dyspnoea and tachypnoea (52/min). Body temperature, hydration, nutritional status, femoral pulses and palpable lymph nodes were normal. Thoracic auscultation revealed increased lung sounds on the right side and decreased lung sounds on the left side. Thoracic percussion was dull on the left side and not well tolerated by the cat.
Radiographic examination of the thorax showed complete soft tissue opacification of the left hemithorax with deviation of the mediastinal structures to the right, and obliteration of thesilhouette of the heart and diaphragm, compatible with the presence of large amounts of fluid or soft tissue in the left hemithorax (Fig 2). Thoracic ultrasonography using the parasternal intercostal approach confirmed a fluid accumulation of moderately corpuscular type in the left hemithorax.
00097-9/unzip/10.1016_S1098-612X(02)00097-9-fig2.png)
Dorso-ventral radiograph of the thorax of the same cat, taken at presentation. Notice the distinct shift of the heart shadow to the right and complete obliteration of the left lung. The cardiac and diaphragmatic silhouettes are obliterated.
In order to alleviate the dyspnoea, a diagnostic and therapeutic thoracocentesis was performed under sonographic guidance. Following localanaesthetic infiltration and disinfection of the chest wall, 100 ml of a sero-sanguineous fluid was obtained. Repeated radiographic and sonographic thoracic examination now showed a marked shift of the mediastinum to the left and collapsed left lung lobes compatible with left-sided atelectasis. No evidence of an intra-thoracic mass was discernible. The right crus of the diaphragm seemed intact; the integrity of the left crus was still impossible to ascertain (Fig 3).
00097-9/unzip/10.1016_S1098-612X(02)00097-9-fig3.png)
Lateral (a) and dorso-ventral (b) radiograph of the thorax of the same cat, taken after thoracocentesis of 100 ml sero-sanguineous fluid. The cardiac silhouette has shifted to left. Complete atelectasis of the left lung lobes is present. The borders of the diaphragm, especially the right crus, and the hear are now partly visible.
A complete blood count (CBC) and serum biochemical analyses were performed. Resultsrevealed mild azotaemia (BUN 12.2 mmol/l; reference range, 3.5 to 8.8 mmol/l), elevation of hepatic transaminases (AST 38 IU/l; reference range, 0 to 22 IU/l; ALT 91 U/l; reference range, 0 to 23 IU/l), lactate dehydrogenase (LDH 194 IU/l; reference range, 0 to 100 IU/l) and creatinine kinase (CK 498 IU/l; reference range, 15 to 130 IU/l). Haematology was unremarkable. Analysis of the pleural fluid based on the serosanguineous aspect, specific gravity 1.017 g/l, RBC 0.13×1012, WBC 4.93×1012 and total protein 1.95 g/dl was consistent with a modified transsudate (Tyler & Cowell 1989, Fossum 2000). Cytological examination showed a rich population of inflammatory cells, almost exclusively degenerated neutrophils, and did not identify any infectious agents.
Primary differential diagnoses were traumatic effusion, diaphragmatic hernia, neoplastic effusion, left lung lobe torsion, pleuritis and pulmonary thromboembolism (Hawkins 2000, Neath et al 2000, Fossum et al 1992). The cat was stabilised with fluids (lactated Ringer solution, IV, at a rate of 2.0 ml/kg body weight, per hour) and closely monitored for oxygen saturation of haemoglobin using pulse oximetry. Over night, the cat became increasingly alert, started eating, maintained oxygen saturation over 90%, but was still tachypnoeic. The next day, the cat was re-evaluated ultrasonographically and, again, massive left-sided thoracic effusion was found. Thoracocentesis was repeated and 200 ml of a macroscopically identical fluid was withdrawn. As the cat continued to do relatively well, it was decided to re-evaluate the thorax the following day as with a pure traumatic lesion a diminution of fluid formation could be expected. Radiographs taken the following morning showed, however, an unchanged picture compared to those at first presentation.
Exploratory thoracotomy was carried out through the left 6th intercostal space. The pleural cavity was filled with large amounts of serosanguineous fluid. In the caudo-dorsal diaphragmatic angle a firm irregular mass of about 4 cm in diameter covered with a thick layer of fibrin adherent to the diaphragm was detected. There was no evidence of lung lobe torsion, but the entire left lung was completely atelectatic, and was covered by a fibrin layer, as was the whole parietal and visceral pleural surface. A rupture of the diaphragm could not be visualised. The thoracotomy was closed in a routine manner and a midline laparotomy was performed. At abdominal exploration, the absence of the left kidney and a small slit-like circumferential tear of about 2 cm at the dorsal attachment of the left lumbar portion of the diaphragm to the psoas minor muscle were noted. Through this opening, part of the mesentery was herniated into the thoracic cavity. All other abdominal organs seemed intact and in place. After enlargement of the hernial ring of the diaphragm, the hernial content, consisting of the left kidney, surrounded by peri-renal fat and covered by fibrin, and one half of the spleen, which had been severed into two pieces, were replaced into the abdomen. Due to the enormous amount of fibrin on the well-vascularised kidney it was impossible to assess the integrity of the ureter. Considering the degree of inflammation and fibrin deposition in the left hemi-thorax, a simple inflammatory reaction due to herniated abdominal organs seemed unlikely. The most likely reason was estimated to be a chemical irritation due to urine leakage. This hypothesis was confirmed by perioperative in-house laboratory evaluation of the thoracic effusion revealing very high levels of urea nitrogen (24.1 mmol/l) and creatinine (472.9 μmol/l). As it was deemed impossible to repair the left urinary pathways, left-sided nephrectomy was performed. Furthermore, diaphragmatic herniorrhaphy usingmattress sutures of polydioxanone 2.0 metric (PDS II, Johnson & Johnson Medical) was accomplished and air was withdrawn from the thoracic cavity using a thoracic drain before tying the last suture. The two parts of the spleen, having sound smooth borders and being well vascularised, were left in the abdomen. Finally, the abdomen was closed in a routine fashion. Postoperative recovery was uneventful. The cat was treated with morphine (Stellorphine, Laboratoires Stella) (0.1 mg/kg body weight, IM) and cephazolin (Cefacidal, Bristol-Myres Squibb Belgium) (20 mg/kg body weight, IV, q 8 h) during surgery and postoperatively for 32 h.
The following day, radiographic examination showed slight thoracic effusion and free air in the abdomen. The left lung lobes were subnormally inflated but showed some pulmonary expansion. The cat was discharged from the hospital with cephalexin (Ceporex, Schering-Plough Animal Health) (15 mg/kg body weight, PO, q 12 h) for 7 days and carprofen (Rimadyl, Pfizer Animal Health) (2.0 mg/kg body weight, PO, q 12 h) for 3 days.
At follow-up examination 10 weeks post operatively, the cat had regained a normal respiratory pattern and normal activity. Radiographic examination revealed a near complete expansion of the left hemithorax with slightly rounded borders of the left diaphragmatic lung lobe. The pulmonary parenchyma appeared normal without any radiographic signs of pleurisy. Proliferative reactive bone formation was observed on the 6th and 7th rib at the site of the thoracotomy. Follow-up CBC and serum biochemical analyses were normal (BUN 7.6 mmol/l; creatinine 87 μmol/l).
Discussion
Diagnosis was not straightforward in this cat, due to lack of evidence of trauma, difficulties in demonstrating the hernia with diagnostic imaging techniques and the unusual organ herniated. Possible reasons for the mild azotemia and the elevation of hepatic enzymes included subclinical dehydration and blunt trauma, however there was no historical evidence for that. Lung lobe torsion secondary to thoracic effusion, although rarely on the left side, remained an important differential diagnosis until exploration (Neath et al 2000). Feline infectious peritonitis was extremely unlikely considering the low protein content of the thoracic effusion, the low plasma globulin content and the unilateral distribution of the fluid. In light of the rapid reformation of the fluid, surgical exploration of the thorax seemed indicated. A thick fibrin layer covered thehernia rendering identification of the hernial content impossible without tissue damage. At laparotomy, however, identification of the hernia was evident. The presence of urine in the thorax explained the mild azotemia noted at first presentation, through pleural re-absorption of the intra-thoracic urine. At this point of time, we were not able to rule out a possible bilateral renal impairment, which also could lead to azotemia. To our knowledge, the only functional test of renal filtration performance capable of distinguishing between the left and right kidney is measurement of glomerular filtration rate by scintigraphy. This test is currently not available in a clinical setting in our country and was therefore not performed on this patient pre-operatively. Re-anastomosis of ureters in cats is associated with an extremely high rate of postoperative obstruction (Bjorling & Christie 1993, Weisse et al 2002). Considering the fact that chronicity of the ureteral rupture was making post-operative stricture formation even more likely, we elected to perform a left-sided nephrectomy, assuming that renal function of the contralateral kidney was normal in this young age cat. At the time of surgery this assumption was based on a blood sample only, without a urine analysis being performed pre-operatively.
Diaphragmatic hernia with prolapse of the kidney is a rare condition. To the author's knowledge, there are only two shortnotes in veterinary literature describing mainly the radiographic findings of this condition in a cat (Rochat & Aronson 1991, Dorey et al 2000). In contrast to the two aforementioned reports where the right kidney was displaced, in our case the left kidney was found intra-thoracically. Dorsally, the diaphragm is attached with its paired lumbar portions to the vertebral bodies of third and fourth lumbar vertebrae. In this location (arcus lumbocostalis), the aorta, right azygos vein, the hemi-azygos vein and the lymphatic duct enter the abdominal cavity, though staying retroperitoneally. The kidney and the surrounding peri-renal fat pad are also retroperitoneal structures (Nickel et al 1982). In the present case, we hypothesise that, due to trauma, the left kidney ‘slipped’ from the retroperitoneal space into the thoracic cavity via a small disruption of the attachment of the dorsal diaphragm. In this movement, the spleen, situated in close proximity was involved and partly herniated as well. During the prolapse, the vascular supply, which reaches the left kidney from more cranially (Nickel et al 1982), was able to follow the movement and to assure vascularisation of the kidney. The ureter however, which leaves the renal pelvis in a caudal direction must have been severely stretched. The presence of the abdominal organs in the thoracic cavity might explain the dyspnoea the cat presented with several weeks previously. A second trauma, such as a jump or fall from a height, possibly led to the rupture of the ureter several weeks later. The urine produced thereafter accumulated in the left hemi-thorax, inducing a chemical fibrosing pleuritis and compression atelectasis of the lung lobes. Urothorax is a very rare condition in small animals and has, to the author's knowledge, never been described before in veterinary medical literature.
Retrospectively, the fact that the second kidney could not be clearly identified on the first radiograph (Fig 1) should have prompted further investigation using deep abdominal palpation, ventro-dorsal or dorso-ventral radiography or thorough abdominal ultrasonography. Intravenous excretory urography could also have been employed to identify the intra-thoracic kidney or to rule out renal agenesis (Christie & Bjorling 1993, Robinson 1965, Pechman 1982). Thoracic ultrasonography using a paraspinal intercostal approach is a further technique that was neglected in this cat. Furthermore, whenever there is any clinical doubt about the origin of a thoracic effusion, biochemical analysis of urinary products easily demonstrates urothorax and should be performed systematically (Tyler & Cowell 1989, Salcedo 1986). In order to assess renal function and to rule out urinary tract infection a urine analysis should also have been performed preoperatively.
Intrathoracic herniation of a kidney has been reported anecdotally in humans. It has been encountered after severe trauma in combination with other injuries and mostly in young individuals (Contamin et al 1977, Teyssier et al 1983, Stylianos et al 1991, Hase et al 1999). Besides that, an acquired form without any evident trauma in the absence of diaphragmatic hernia has been described (Suarez & de Jesus 1998). Once the challenging diagnosis has been established, the appropriate treatment options are often easier to envisage and can lead to an excellent outcome.