BUTLER, ERIN L. MD; COX, SUSAN M. MD; EBERTS, ERIC G. MD; CUNNINGHAM, F. GARY MD
Department of Obstetrics and Gynecology, University of Texas Southwestern Medical Center, Dallas, Texas.
Address reprint requests to: Erin L. Butler, MD, Department of Obstetrics and Gynecology, University of Texas, Southwestern Medical Center, 5323 Harry Hines Boulevard, Dallas, TX 75390-9032
Received December 20, 1999. Received in revised form May 15, 2000. Accepted May 26, 2000.
Objective: To review our experiences with diagnosis and management of symptomatic nephrolithiasis complicating pregnancy and to ascertain the efficacy of renal sonography for initial diagnosis compared with plain x-rays or single-shot intravenous pyelography.
Methods: Perinatal outcomes were evaluated for all pregnant women admitted to Parkland Hospital for nephrolithiasis from 1986 to 1999. Diagnostic studies and management of nephrolithiasis were also evaluated.
Results: During the 13-year study period, 57 pregnant women had 73 admissions for symptomatic nephrolithiasis. Symptomatic nephrolithiasis complicated 1 in 3300 (0.03%) deliveries at our institution. Only 12 women (20%) had a history of renal calculi. Mean gestational age at diagnosis was 23 weeks. Imaging techniques included renal ultrasonography, plain abdominal x-ray, and single-shot intravenous pyelography. Calculi were visualized in 21 of 35 (60%) renal ultrasonographic examinations and 4 of 7 (57%) abdominal x-ray studies when these were performed as the initial test. In contrast, urolithiasis was discovered in 13 of 14 (93%) instances in which intravenous pyelography was performed as the initial diagnostic test. When sonography was negative (n = 14), renal calculi were confirmed by single-shot intravenous pyelography (n = 8). Although 43 of 57 (75%) of symptomatic episodes responded to conservative management, 10 women required ureteral stents, 3 needed percutaneous nephrostomy tubes, and 2 underwent ureteral laser lithotripsy for resolution.
Conclusion: Although the convenience and safety of ultrasonography to initially diagnose nephrolithiasis are indisputable, 40% of calculi were missed when this method alone was used. Thus, if nephrolithiasis is still suspected clinically despite ultrasonographic findings, single-shot pyelography is recommended.
After a review of 14 series published over a 50-year period, Hendricks et al1 reported the average incidence of symptomatic nephrolithiasis to be 1 in 1240 pregnancies (0.08%). This incidence is remarkably similar to that cited for nonpregnant women of childbearing age.2 More recently, the Calcium for Prevention of Preeclampsia Group performed careful antepartum screening of 4589 pregnant women and found that approximately 1 in 1100 had urolithiasis; however, only one of four such women had symptoms.3 Although pregnancy does not seem to alter the incidence of symptomatic nephrolithiasis, there are reasons why it might be expected to do so. For example, pregnancy-induced urinary stasis and hypercalciuria predispose to calculus formation. Conversely, pregnant women with calculi may have fewer symptoms because ureteral dilatation permits easier passage of calculi.
Diagnosis of nephrolithiasis in pregnancy is problematic because of physiologic urinary tract dilatation. The difficulty is compounded by an inherent reluctance to perform x-rays or invasive diagnostic procedures in pregnant women. Four reports in the last decade have described urolithiasis in a total of 67 pregnant patients.1,4–6 The largest series included 29 women but did not address diagnostic approaches during pregnancy.5
We describe our experience with diagnosis and management of symptomatic nephrolithiasis in 57 pregnant women. These women were treated in a consistent manner, and their evaluation, management, and pregnancy outcomes are described.
Materials and Methods
Computerized discharge diagnoses were used to identify pregnancies complicated by symptomatic nephrolithiasis at Parkland Hospital in the 13-year period from April 1986 through March 1999. Fifty-seven women were identified; they usually presented with abdominal or flank pain, frequently accompanied by hematuria. All women suspected of having renal calculi underwent renal ultrasonography, single-shot intravenous pyelography, or plain abdominal x-ray. Our preferred method for initial imaging is renal ultrasonography. Single-shot pyelography, in which a single radiograph is obtained 30 minutes after introduction of contrast medium, was occasionally used for initial evaluation but was almost always used in women with suspected nephrolithiasis who had negative findings on ultrasonography. In all instances, the diagnosis of urinary calculi was confirmed by either imaging studies or spontaneous passage of a calculus.
The women were managed using conservative methods that included analgesia, intravenous crystalloid for hydration and diuresis, and antimicrobial therapy for concomitant infection. Invasive procedures, such as stent placement, percutaneous nephrostomy, and ureteral laser lithotripsy, were reserved for patients with continued symptoms and evidence of ongoing obstruction. The clinical courses of these women were analyzed to ascertain their clinical presentation, efficacy of diagnostic imaging methods, treatment and response, and pregnancy outcomes.
Over the 13-year period, 57 pregnant women required 73 admissions for nephrolithiasis. During this same period, 186,271 women delivered; thus, the incidence of symptomatic nephrolithiasis was 0.03%, or 1 per 3300 deliveries. The demographic characteristics of these 57 women are shown in Table 1. Only 12 women had a history of nephrolithiasis. Nephrolithiasis was diagnosed in 13 women (20%) in the first trimester, 22 (40%) in the second trimester, and 23 (40%) in the third trimester.
The most common clinical findings are shown in Table 2. Confidence intervals (CI) are presented for percentages calculated by using the method of exact binomial proportions. Abdominal or flank pain was the impetus for presentation in 89% of women. About one third of the 57 women had nausea, dysuria, or gross hematuria. The most common laboratory findings were bacteriuria (81%) and hematuria (75%).
At the time of initial presentation, nephrolithiasis was correctly diagnosed in 35 (61%) women, of whom 12 had concomitant pyelonephritis. In another 12 (20%) of the 57 women, pyelonephritis was diagnosed; however, continued fever or pain despite 72 hours of intravenous antimicrobial therapy prompted investigation that disclosed renal calculi. In the remaining 10 women (18%), admission diagnoses included term labor, hematuria of unknown cause, hyperemesis, or suspected appendicitis; however, clinical observation prompted diagnosis of nephrolithiasis.
Diagnostic confirmation of nephrolithiasis was obtained by sonography, plain abdominal x-ray, or single-shot intravenous pyelography. In some women, multiple diagnostic procedures were necessary to confirm the diagnosis. One woman passed a calculus before imaging studies were performed. Renal ultrasonography was used as the primary diagnostic tool in 35 of 56 women, but calculi were visualized in only 21 (60%; 95% CI 47%, 76%) of these women. Of the 14 women with negative findings on ultrasonography, calculi were detected by single-shot pyelography in 8 and by urine straining for calculi in 6. Single-shot pyelography was used in another 14 women, and calculi were visualized in 13 (93%; 95% CI 66%, 100%). In 7 women evaluated initially by single plain abdominal x-ray, calculi were correctly diagnosed in 4 (57%; 95% CI 18%, 90%). Calculi in the 3 women with negative results on single plain abdominal x-ray were diagnosed by using single-shot pyelography.
Treatment of nephrolithiasis included analgesia for pain, intravenous crystalloid hydration, and antimicrobial treatment for concomitant infection (the latter was given to half of the patients). This conservative management strategy resulted in resolution of symptoms in 43 (75%) of the 57 women. In the 14 women (25%) whose symptoms did not abate, invasive procedures were required. A ureteral stent was placed in 10 women and percutaneous nephrostomy tubes were placed in 3. Two women underwent ureteral laser lithotripsy in the first trimester; one of the two women required this intervention after placement of a percutaneous nephrostomy tube.
The most common complication was recurrent symptomatic nephrolithiasis. An extreme example is one woman who had a percutaneous nephrostomy tube placed at 23 weeks and required five additional hospitalizations for complications related to the nephrostomy tube and recurrent symptoms. One woman who underwent numerous radiologic procedures with an undiagnosed early pregnancy was calculated to have received a fetal dose of 5 to 6 rads without resolution of disease. She chose pregnancy termination and underwent definitive therapy that was calculated to require a cumulative dose of 10 rads. Two women spontaneously aborted at 10 and 19 weeks; both instances were unrelated temporally to admissions for nephrolithiasis. The 46 women who delivered at Parkland Hospital had a mean gestational age at delivery of 40 weeks (range 30 to 42 weeks), and mean birth weight was 3075 g (range 1820–3965 g). Delivery was vaginal in 35 women and by cesarean section in 11 (22%). The incidence of pregnancy-induced hypertension was 16%.
The incidence of symptomatic nephrolithiasis in our population was 1 in 3300 deliveries. This incidence is lower than that in the review by Hendricks et al,1 but it is probably representative of a nonreferred population. The incidence is remarkably similar to that in the study reported by the Calcium for Preeclampsia Prevention group.3 In this latter study, careful surveillance methods were used to document an incidence of symptomatic nephrolithiasis of 1 in 4600 pregnancies.
We found that renal ultrasonography had a high rate of false-negative results. Only 60% of cases were confirmed by this method when it was used initially. This rate is similar to the 40% to 50% that others have reported.1,4 In 24 of 25 women (96%), single-shot intravenous pyelography was confirmatory. Calculi were visualized on abdominal flat-plate x-rays in 4 of 7 studies, a rate similar to that obtained with ultrasonography. Of note, a single x-ray delivers only about 50 mrad to the fetus.7,8
Three-fourths of the 57 women responded favorably to conservative management with intravenous hydration, pain relief, and antimicrobial agents for concomitant infection. In the 25% who did not respond, invasive treatment was undertaken for renal obstruction confirmed by pyelography or persistent symptoms. Although other investigators1,4–6 have reported obstruction in 30–50% of pregnant women with symptomatic calculi, the lower rate that we describe can probably be attributed to our nonreferred patient population.
Pregnancy outcome was not appreciably worsened because of symptomatic nephrolithiasis. There were two seemingly unrelated spontaneous abortions and one induced abortion after excessive exposure to diagnostic radiation. Of 46 women who delivered at Parkland Hospital, only one was delivered preterm (30 weeks); that patient had been treated for nephrolithiasis at 9 weeks. The cesarean delivery rate of 22%, including six repeat procedures, is similar to our overall rate of about 20%. Similarly, the overall incidence of 16% for pregnancy-induced hypertension is not excessive.
We conclude that ultrasonographic evaluation of pregnant women with suspected calculi is a reasonable diagnostic procedure to use initially because of its safety and convenience. If nephrolithiasis is not confirmed by this method, however, and if symptoms suggesting the presence of renal calculi persist, single-shot intravenous pyelography should be performed.
1. Hendricks SK, Ross SO, Krieger JN. An algorithm for diagnosis and therapy of management and complications of urolithiasis during pregnancy. Surg Gynecol Obstet 1991;172:49–54.
2. Coe FL, Parks JH, Lindheimer MD. Nephrolithiasis during pregnancy. N Engl J Med 1978;298:324–6.
3. Levine RJ, Hauth JC, Curet LB, Sibai BM, Catalano PM, Morris CD, et al. Trial of calcium to prevent preeclampsia. N Engl J Med 1997;337:69–76.
4. Haller H, Mozetic V, Topljak-Polic D, Randic L, Fuckar Z. Management and complications of urolithiasis during pregnancy. Int J Gynaecol Obstet 1993;40:135–9.
5. Denstedt JD, Razvi H. Management of urinary calculi during pregnancy. J Urol 1992;148:1072–5.
6. Buchholz NP, Biyabani R, Sulaiman MN, Talati J. Urolithiasis in pregnancy: A clinical challenge. Eur J Obstet Gynecol Reprod Biol 1998;80:25–9.
7. Swanson SK, Heilman RL, Eversman WG. Urinary tract stones in pregnancy. Surg Clin North Am 1995;75:123–42.
8. Cunningham FG, MacDonald PC, Gant NF, Leveno KJ, Gilstrap LC III, Hankins GDV, et al. Williams obstetrics. 20th ed. Stamford, Connecticut: Appleton & Lange, 1997.