The management of symptomatic hydronephrosis in pregnancy
*Correspondence: Dr O. Fainaru, Department of Obstetrics and Gynecology, Lis Maternity Hospital, Tel Aviv Sourasky Medical Center, 6 Weizmann Street, Tel Aviv 64239, Israel.
Objective To present and to evaluate the conservative and surgical management of hydronephrosis in pregnancy.
Design Retrospective analysis of an interventional cohort.
Setting A tertiary maternity ward.
Sample Of the 30,552 women delivering in our institution between January 1998 and June 2001, 56 women (0.2%) were admitted to the prenatal care unit due to symptomatic hydronephrosis.
Methods For each patient, renal sonography, urinalysis, serum creatinine levels, white blood cell counts and urine culture were obtained. All patients were treated conservatively by analgesics, intravenous fluids and antibiotics. Failure of these measures: non-resolving infection, deteriorating renal function, absence of Doppler sonographic evidence of ureteral flow or intractable pain led to instrumental drainage of the affected kidney. A double pigtail polyurethane ureteric stent was passed under cystoscopic vision and sonographic guidance.
Main outcome measures Resolution of clinical symptoms.
Results Conservative treatment led to resolution in 52 women (92.9%), whereas four women (7.1%) failed to respond and were treated successfully and without complications by pigtail insertion. Early induction of labour was unnecessary and good perinatal outcome was the rule.
Conclusions Although the vast majority of cases of symptomatic hydronephrosis in pregnancy may be treated conservatively, pigtail insertion is an efficient and safe modality for the rare patient with refractory symptoms.
During pregnancy, mild hydronephrosis is considered a normal phenomenon and may be present in up to 90% of pregnancies1–3. Dilation is usually more pronounced on the right kidney4,5 in primigravidas6 and after mid-pregnancy6. This dilation disappears a few weeks after birth7. Acceptable explanations for this phenomenon are compression of the gravid uterus on the ureters and the smooth muscle relaxing influence of progesterone1,8. The predisposition for the right side may be explained by the dextrorotation of the uterus and the relative protection of the left ureter provided by the sigmoid colon. Increased diuresis, small stones or other unrecognised factors may cause decompensation of ureteral function, progressing to symptomatic acute hydronephrosis9. We performed this study to evaluate the role of the different treatment measures practiced in our institution for the treatment of acute symptomatic hydronephrosis in pregnancy (i.e. medical treatment vs sonographic guided drainage).
Of the 30,552 women delivering in our institution between January 1998 and June 2001 (3.5 years), 56 women were admitted to the prenatal care unit for symptomatic hydronephrosis (flank and loin pain, pyelonephritis or urinary obstruction). For each woman, renal sonography was performed. Only one patient had evidence of urinary calculi. Hydronephrosis was graded according to maximal calyceal diameter, as detailed by Zwergel et al.9: 5–10, 10–15 and >15 mm were considered mild, moderate and severe hydronephrosis, respectively. Urinalysis, serum creatinine levels, white blood cell counts and urine culture were also performed. All patients were treated by conservative measures: analgesics and intravenous fluids. Intravenous antibiotics (cefuroxime and gentamicin) were added on signs of infection (i.e. fever, leucocytosis).
On failure of conservative measures: (1) non-resolving signs of infection after 48 hours, (2) deteriorating renal function (i.e. increase in serum creatinine or BUN values), (3) absence of Doppler sonographic evidence of ureteral flow (‘jet sign’) indicating ureteric obstruction or (4) intractable pain, drainage was achieved surgically. A double pigtail polyurethane uretric stent (Cook Urological, Spencer, Indiana, USA) was passed under cystoscopic vision and sonographic guidance. This method allows internal drainage from the ureteropelvic junction to the urinary bladder. The procedure was done under intravenous sedation.
An acute symptomatic hydronephrosis of pregnancy was diagnosed in 56/30,552 (0.2%) pregnant women delivering in our institution between January 1998 and June 2001.
The mean demographic and clinical data for the conservatively treated patients and absolute data for the patients treated with pigtail insertion are summarised in Table 1.
Table 1. Patient demographic and clinical data [mean (SD)].
|Age (years)||26 (5)||29||20||30||30|
|GA (weeks)||27.6 (5.6)||28||23||28||26|
|GA at delivery (weeks)||39.2 (1.4)||39||39||38||34|
|Serum BUN (mmol/L)§||2.1 (0.8)||9.3||2.5||10.7||3.6|
|Serum creatinine(μmol/L)§||60.8 (13.3)||123.8||70.7||132.6||88.4|
|WBC (×103/μL)||12.1 (0.4)||22||9||8.5||11.7|
|Hydronephrosis grade|| || || || || |
|2||19.2%|| || || || |
|3||7.6%|| || || || |
|Hospitalization (days)||5.3 (1.6)||8||10||16||11|
Urine culture was positive for bacteria in 48% of patients. Of the culture positive women, the specific bacteria were: Escherichia coli (62.9%), Klebsiella pneumonia (14.8%), Enterococcus faecalis (11.1%), mixed cultures (3.8%) and Streptococcus agalctiae (1.9%).
Conservative treatment led to resolution of symptoms and signs in 52/56 women (92.9%), whereas 4/56 women (7.1%) failed to respond to these measures and were treated by double pigtail ureteral stent insertion. All women were delivered at term (38.5–39.5 weeks, 95% confidence interval [CI]), and had a good perinatal outcome.
The specific indications for pigtail insertion were (Table 1): impairment of renal function as reflected by increasing serum creatinine (two cases), non-resolving pyelonephritis with the development of systemic inflammatory response (i.e. respiratory distress), absence of the ‘jet sign’ in Doppler sonograpy and intractable pain. Insertion of the stent was without any complications and was followed by rapid resolution of pain, fever, leucocytosis, respiratory symptoms and improvement of renal function. Pigtail stents were inserted between days two and five of hospitalisation and patients were discharged within 8–16 days. Three of these women had normal deliveries and good perinatal outcomes, whereas in one woman, labour was induced at 34 weeks of gestation due to a psychiatric indication. In all cases, ureteric stents were removed without complications within four to six weeks after delivery.
The physiologic dilation of the urinary collecting system in pregnant women may become symptomatic, and if left untreated can progress to life threatening infection which may endanger the mother and the fetus5. We find that sonography with the assistance of Doppler is suitable for evaluation of hydronephrosis grade and ureteric function. Intravenous pyelography was unnecessary in all cases. The relatively low frequency of acute hydronephrosis in pregnancy (0.2%) over a 42-month period observed in our study, when compared with other reports (3%6), may be a result of appropriate outpatient care and the low admission rate for this condition in our institution (restricted for women presenting with acute pyelonephritis or suffering from unrelenting pain caused by ‘overdistention’4). Our results accord with previous findings5,6,10 of right predominance of hydronephrosis in pregnancy and of higher incidence of this condition in primiparas. Furthermore, as expected6, we observed that the majority of the cases were diagnosed after mid-pregnancy. In the majority of our patients (92.9%), symptoms resolved after two to five days of conservative management (analgesia, hydration and antibiotics). This agrees with the incidence reported in previous studies (i.e. 73%9, 94%5). In the few cases when these measures failed, the relatively simple and safe method of double pigtail stent insertion was successful.
Although this procedure carries the risk for several complications such as catheter migration, ascending pyelonephritis caused by vesico-urethral reflux and stone formation9,11,12, we observed no such complications in our cases. Aside from minor transient flank discomfort, the procedure was uneventful. Nevertheless, the low incidence of complications in our study may be due to the small number of procedures performed.
Notably, none of the patients needed previously practiced more aggressive modalities, such as percutaneous nephrostomy13,14. The introduction of indwelling ureteric catheters (draining the ureter all the way through the urethra) is being abandoned because of poor patient compliance with this instrumentation7.
Fifty-five out of our 56 patients carried their pregnancies to term (in one, labour was induced at 34 weeks for reasons non-relevant to the urinary tract). Early induction of labour for the relief of urinary symptoms or signs was unnecessary in all cases. Good perinatal outcome was the rule in all our cases. However, hydronephrosis caused by ureteral obstruction has been reported to cause pregnancy complications (such as pre-eclampsia)15.
Most acute non-physiological hydronephroses during pregnancy are caused by renal calculi, but overall this phenomenon is rare (0.04%)16. In our study, only one patient had evidence of a small pelvic stone, and this patient was treated successfully with conservative measures. None of our women had a past history of other urologic conditions such as vesico-ureteric reflux.
Although our study involved only four cases of pigtail insertion, we can conclude that this is an efficient and safe modality for the treatment of the relatively rare patient with refractory signs and symptoms of hydronephrosis of pregnancy. This measure leads to effective drainage and prompt resolution of this condition. Nevertheless, the vast majority of cases of acute hydronephrosis in pregnancy may be successfully treated conservatively.