Ketonuria is not associated with hyperemesis gravidarum disease severity

To assess the association between ketonuria and hyperemesis gravidarum (HG) disease severity.

We included pregnant women hospitalised for HG who participated in the Maternal and Offspring outcomes after Treatment of HyperEmesis by Refeeding (MOTHER) trial and women who were eligible, chose not to be randomised and agreed to participate in the observational cohort. Between October 2013 and March 2016, in 19 hospitals in the Netherlands, women hospitalised for HG were approached for study participation. The presence of ketonuria was not required for study entry. Ketonuria was measured at hospital admission with a dipstick, which distinguishes 5 categories: negative and 1+ through 4 + . The outcome measures were multiple measures of HG disease severity at different time points: 1) At hospital admission (study entry): severity of nausea and vomiting, quality of life and weight change compared to pre-pregnancy weight, 2) One week after hospital admission: severity of nausea and vomiting, quality of life and weight change compared to admission, 3) Duration of index hospital admission and readmission for HG at any time point RESULTS: 215 women where included. Ketonuria was not associated with severity of nausea and vomiting, quality of life or weight loss at hospital admission, nor was the degree of ketonuria at admission associated with any of the outcomes 1 week after hospital admission. The degree of ketonuria was also not associated with the number of readmissions. However, women with a higher degree of ketonuria had a statistically significant longer duration of hospital stay (per 1+ ketonuria, difference: 0.27 days, 95 % CI: 0.05 to 0.48).

There was no association between the degree of ketonuria at admission and severity of symptoms, quality of life, maternal weight loss, or number of readmissions, suggesting that ketonuria provides no information about disease severity or disease course. Despite this, women with a higher degree of ketonuria at admission were hospitalised for longer. This could suggest that health care professionals base length of hospital stay on the degree of ketonuria. Based on the lack of association between ketonuria and disease severity, we suggest it has no additional value in the clinical management of HG.