Letters to the Editor


  • Dr Ivy HY Chan,

  • Dr Wendy WM Lam,

  • Dr Kenneth KY Wong,

  • Prof. Paul KH Tam

    1. Chair of Pediatric Surgery
      Department of Surgery
      The University of Hong Kong
      Hong Kong
    Search for more papers by this author

20 August 2009

Dear Editor,

Renal pelvis haematoma causing pelviureteric obstruction: a first case of Antopol-Goldman lesion in a neonate

Neonatal hydronephrosis is a relatively common problem. Most are diagnosed by routine antenatal ultrasound screening, with an estimated frequency of between 0.6% and 1.4%.1,2 The most common cause is pelvi-ureteric junction obstruction (PUJO), which accounts for more than 60% of the neonatal hydronephrosis with genuine pathology. Vesicoureteric reflux, vesico-ureteric junction obstruction are the other causes.3 The pathology for PUJO is usually due to an intrinsic stenosis, followed less commonly by extrinsic compression.4

In adults, Antopol and Goldman first described a subepithelial haematoma of renal pelvis causing urinary obstruction in 1948.5 In subsequent publications by others, haematuria was reported to be the usual presenting symptom, with radiological investigations showing a renal mass. Unfortunately, many of these lesions were mistaken as renal neoplasms and the patients were subjected to unnecessary nephrectomies.6,7 Here we report a preterm baby who presented with PUJO caused by a haematoma. We believe that this is the first reported case of Antopol–Goldman lesion in the paediatric population.

Our patient was a girl born prematurely at 29 weeks of gestation by spontaneous vaginal delivery. Antenatal ultrasound did not show any abnormality.

The premature labour was due to antepartum haemorrhage from placental previa. One dose of antenatal steroid was given before delivery. Her birthweight was 1.2 kg.

She was cyanotic and apnoeic at birth and required intubation and mechanical ventilation, and was given two doses of surfactant. Our patient was found to have renal impairment since day 1 of life, with the level of creatinine climbing to 140 micromol/L on day 5 of life. There was no oliguria nor macroscopic haematuria. She was treated as having dehydration with intravenous fluid resuscitation and her renal function slowly improved subsequently. No ultrasound was done at this point because of improving renal function.

Abdominal distension was noted on day 11. Abdominal X-ray showed evidence of pneumatosis intestinalis, which was compatible with necrotising enterocolitis. This was treated with conservatively and our patient made an uneventful recovery. Feeding was resumed after 5 days. On day 21 of life, physical examination revealed a right upper quadrant mass measuring around 2 cm in diameter. Ultrasonogram of the abdomen showed a grossly hydronephrotic right kidney measuring 5.2 × 4.8 × 3.2 cm. A solid mass (2.1 × 1.4 × 1.8 cm) was noted at right renal hilum and extending into right ureter (Fig. 1a). There was no significant vascular flow demonstrated in the lesion. The left kidney was unremarkable. Routine blood tests, including the clotting profile, were normal. Computed tomography (CT) of the abdomen was performed to ascertain the nature of the mass at right renal hilum. This showed a mildly hyperdense lesion inside the dilated right renal pelvis causing hydronephrosis. No contrast enhancement was noted inside the mass (Fig. 1b). A radiological diagnosis of renal pelvis haematoma causing PUJO was made. As there was significant PUJO with evidence of cortical thinning, right-sided percutaneous nephrostomy was performed. The antegrade pyelogram showed complete obliteration of right renal pelvis. The initial urine drained from the nephrostomy contained clots, which was compatible with the initial diagnosis.

Figure 1.

Imaging studies of the right kidney of the patient demonstrating the obstructing lesion at the right renal pelvis. (a) Ultrasound showing gross hydronephrosis and an echogenic mass lesion inside the renal pelvis (arrow). (b) Post-contrast oblique coronal computed tomography showing the mass lesion causing outflow obstruction (arrow).

Serial ultrasound performed subsequently showed decreasing size of the mass lesion, suggestive of resolving haematoma. A repeated right-sided antegrade pyelogram was done 1 month later. This showed prompt contrast drainage to urinary bladder and no intraluminal filling defect could be detected (Fig. 2). The nephrostomy tube was removed and the patient was discharged 2 weeks later. Reassessment ultrasound at 3 months of age showed only mildly dilated right renal pelvis, with an otherwise normal urinary system. No mass lesion was detected. Renal MAG3 scan was also performed at 3 months of age. It showed satisfactory function and perfusion over both kidneys with no evidence of obstruction.

Figure 2.

Right antegrade pyelogram showing contrast drainage to the urinary bladder with no evidence of intraluminal lesion.

In the paediatric population, the most common cause of PUJO is due to intrinsic narrowing. Intraluminal obstruction is rare in neonates, with only a few reports found with acute renal failure caused by obstructive candidal bezoars.8,9 Although the Antopol–Goldman lesion was first reported in 1948, it remains a rare entity in adults and has never been reported in neonates. Most of the affected patients presented with haematuria and filling defects in the renal pelvis on radiological investigations. Because of the possibility of renal cell carcinoma, most adults required nephrectomy before coming to the correct diagnosis. Although some patients did report a history of minor trauma before development of this lesion, there has never been a definite cause that can be attributed to this pathology.

In our case, the echogenic mass lesion detected on ultrasonogram showed that it was not arising from the adjacent renal tissue. Furthermore, there was no Doppler flow detected inside the lesion nor was there contrast enhancement on CT to suggest the presence of tumour blood vessels. Reviewing the history in our patient, we believe that the stormy course soon after birth, together with deteriorating renal function, would suggest the possibility of the haematoma having been present since birth. Since renal cell carcinoma was extremely rare in neonates, this diagnosis was not considered and a conservative approach adopted. We are not able to provide a definite explanation for the cause of the haematoma in our patient, although premature labour and antepartum haemorrhage could have contributed towards the formation of a haematoma. In conclusion, we report here the first case of an Antopol–Goldman lesion in a neonate and hope that this case can raise the awareness of clinicians.