Nephrolithiasis and Pregnancy: Has the Incidence Been Rising?

Introduction

The incidence of urolithiasis has been steadily rising over recent decades. ^{1 –4} This increase in stone disease has been seen in both genders and across all ages, but has been quite pronounced in women. ^{5 –9} Diet, environmental changes, and increasing prevalence of comorbid conditions, such as diabetes and obesity, are thought to be contributing factors for the rise in kidney stones. ^{1,9 –11} Whereas incidence of nephrolithiasis appears to be rising across all populations, particularly females, the pregnant population has not been specifically studied and the trend in this cohort is not known. When previously studied, the incidence of stones in pregnant women was comparable to their nonpregnant counterparts. ¹² As such, with the generally rising stone incidence, one would expect a comparable increase in the incidence of stones during pregnancy. We aimed to define the change in incidence of urolithiasis in pregnant women utilizing international classification of diseases, ninth revision (ICD-9) codes.

Materials and Methods

After institutional review board approval, hospital data from a single tertiary referral center were evaluated for ICD-9 codes for pregnancy and urolithiasis between January 1991 and December 2011. Data were able to be obtained from emergency room visits, operative procedures, and hospital admissions. ICD-9 codes for pregnancy included 640–648, V22.0, V22.1, and V22.2. The ICD-9 codes for urolithiasis included were 592.0, 592.1, and 592.9.

The change in incidence for pregnancy, stones, and the combination of both was examined over the entire time period. Data were also subdivided into 5- and 10-year time increments and compared with each other using the two-tailed student's t-test.

Results

Over the 21-year period, 204,034 pregnant females, 3262 patients with stones, and 876 pregnant patients with stones were treated at our tertiary care referral center (Table 1).

The incidence of stones, pregnancy, and stones in pregnant females from the initial 10 years (1991–2000) were compared with the last 11 years (2001–2011). During these time periods, there was an increase in the total number of stones diagnosed from 778 to 2484 (p=0.004). Pregnancy rates at our institution did not significantly differ over time (94,674 vs. 109,360 patients, p=0.346). The incidence of pregnant females diagnosed with stones also did not statistically significantly change during these two time periods (362 vs. 514, p=0.125; Table 2).

To further define change over time, the data were subdivided into 5-year time periods (1991–1995, 1996–2000, 2001–2005, 2006–2010). Over time, there was a statistically significant increase in overall patients with stones diagnosed; this was first noted between 1996–2000 and 2001–2005 and continued into the last 5-year time period studied as well. There was an overall decrease in pregnant patients seen between the second and third time increments, with a significant increase during the last 5 years of the study. There was no significant change in pregnant patients with stones during these 5-year periods (Table 3).

Finally, looking at the years with the largest absolute increase in stone diagnosis and in pregnant patients with stones, we compared data from 2006–2008 to 2009–2011. Again noted was an increase in the total patients diagnosed with a stone (by 2.1 times) with no change in the number of pregnant patients evaluated. During this time period, there was a statistically significant increase in the number of pregnant patients diagnosed with stones by 1.65 times (124 to 205, p=0.007; Table 4).

Worth noting, the last 12 years of total acute adult admissions were also available for review. Total acute adult admissions increased by 1.7 since 2000. During this same time period, stone diagnosis increased by 3 times and total stone diagnosis during pregnancy increased by 1.7 times.

Discussion

Over the last decade, the incidence of nephrolithiasis has been increasing. ^{1 –4} This increase has been even more pronounced among women. ^8,9 Whereas there has been no difference in the prevalence of stones in pregnancy compared with the general population in the past, time trends have not been studied in this population. To our knowledge, our study is the first to report a change in kidney stone diagnosis during pregnancy over time. In contradistinction to the rest of the population, we did not see an overall rise in stone diagnosis among the pregnant cohort studied over a 2-decade period. Whereas there was a slight increase in the last 3 years, this is not consistent with other studies in the general population showing increasing incidence over the last 10–15 years. ^{1 –4} Whereas diet, environmental changes, and increasing comorbidities are among the common hypotheses for the rising rates of nephrolithiasis, the lack of a corresponding increase in stones in the pregnant population suggests that there may be a fundamental difference in the pathophysiology of stone formation in this cohort that has not been yet identified.

There are some known differences between nephrolithiasis during pregnancy and the nonpregnant female stone former. Ross et al. identified that the stone composition was predominantly calcium phosphate at 74% with only 26% being calcium oxalate. This is in contrast to the nonpregnant stone former, where nearly 75% of stones are of calcium oxalate composition. ¹³ Whereas the difference is thought to be pH related, the exact reasons for this variation remain unclear and may play a role in why stone disease during pregnancy has remained stable.

Another difference between the pregnant and general populations is age. Pregnant females are younger patients than the whole of stone formers studied. Whereas the rise in stone incidence has been partially attributed to hypertension, diabetes, and other comorbidities, these conditions may be lower in this younger cohort. ^{9 –11} Obstetric literature, however, has shown an increase in age and subsequent comorbidities like obesity, hypertension, and diabetes. ^14,15 In addition, Rosenberg et al., found that pregnant stone formers were more likely to be obese, hypertensive, and have gestational diabetes. ¹⁶

There are several limitations to our present study. These data are from a single institution. Additionally, it is a large tertiary referral center, which is routinely referred complicated obstetrics and endourologic patients and may not be a true representation of the general population. Whereas one would predict a possible overestimation of stone diagnosis among the pregnant cohort in this setting, multi-institutional data across different regions would help strengthen this study's conclusions. Furthermore, while our particular region is not in the traditional stone belt, we did see an increase in the total patients with stones over the last 10 years, as expected, so this does not fully explain the lack of increase in the pregnant cohort. ^1,2

If patients did not have a procedure performed, were not admitted, or were not seen in the emergency room, their data were not captured. However, these data should have been equally omitted in both the total stone population and the pregnant cohort. Whereas we saw the expected increase in the total number of patients with stones, there was no comparable rise in the pregnant cohort.

The largest limitation of our study is the inherent problems that accompany the use of an administrative database. Mis-coding can result in overestimation or underestimation of diagnostic rates. In fact, the estimated incidence of stones during pregnancy in our study of 0.4% suggests an overestimation, as previously reported rates are lower at 0.05%. ¹⁶ Mis-coding in the same setting, however, would be expected to be uniform over time. Additionally, administrative coding for nephrolithiasis has been previously validated, suggesting that our findings are reliable. ¹⁷

Our findings are thought provoking, and while it remains unclear why the incidence of stones in pregnancy has not increased as it has in nonpregnant stone formers, further study is needed with multi-institutional studies to confirm our results, particularly in the traditional stone-belt regions, where the overall incidence is much higher.

Conclusion

The rate of nephrolithiasis in pregnant patients over a 2-decade period remained stable. This is in contradistinction to the general population, where stone incidence is rising. Further research is warranted to determine why nephrolithiasis in the pregnant population is not increasing. Larger, multi-institutional studies are needed to validate our results.