Introduction of a protocol for the hospital management of hyperemesis gravidarum patients

L. Eberhardt, C. Stubsten, M. Sisson
Northside Hospital, Atlanta, Georgia, U.S.A.

A multidisciplinary team of physicians, nurses, dietitians, and pharmacists developed a protocol to standardize and improve the treatment of hyperemesis gravidarum (HG). The goals of the team were to reduce length of stay, patient charges, use of total parenteral nutrition (TPN) and readmission rate. The protocol includes an algorithm of antiemetic therapy (metoclopramide, promethazine, droperidol, and ondansetron) in addition to standard physician orders. On admission, the patient receives intravenous fluid, antiemetics, multivitamins and thiamine, consultations with appropriate disciplines, and allowed preferred foods. Control of symptoms is defined by a reduction in nausea and attainment of caloric intake goals. Nasogastric tube feeding (TF) is initiated when nausea is not controlled in 48 hours or the patient is determined as nutritionally high risk. The patient is evaluated for alternate therapy when tube feeding is not tolerated or indicated. Data were compared between a 3-month period of May-July, 1997 (pre-protocol) and May-July, 1998 (post-protocol) resulting in 39 admissions vs. 33 admissions, respectively. During the post-protocol period, length of stay declined by 11% and average pharmacy charges were reduced by 23%. Clinically, the use of TPN has been reduced by 91%, coupled with an 84% increase in the use of TF. Four patients experienced multiple admissions in 1997 vs. eight in 1998. Nine of the 12 patients with multiple admissions ultimately required TPN or TF. The increase in readmissions was unexpected, possibly associated with differing levels of HG severity between the two groups. Preliminary data indicates that the use of a standardized, multidisciplinary protocol may reduce length of stay and patient charges in this population and that TF can be an effective, less expensive, non-invasive alternative to TPN. While these trends are positive, group differences were not statistically significant. This may be due to small sample size and large variance between subjects. Future study will include a larger experimental period and investigate effects of the protocol on maternal and fetal outcome.

Back to home