Nutrition support for bone marrow transplant patients.

Abstract

Bone marrow transplantation involves the administration of toxic chemotherapy and infusion of marrow cells. After treatment, patients can develop a poor appetite, mucositis and gastrointestinal failure, leading to malnutrition. To prevent this, parenteral nutrition (PN) support is the first choice but is associated with an increased risk of infection. Enteral nutrition (EN) is an alternative, as is the addition of substrates e.g. glutamine to enteral and parenteral solutions. However, the relative effectiveness of these treatments is not clear. To determine the efficacy of EN or PN support for patients receiving a bone marrow transplant. Trials were identified by searching the Cochrane Library (Issue 4, 2000 ), MEDLINE (1966-2000), EMBASE (1988-2000) and CINAHL (1982-2000 ). Reference lists of identified trials and conference proceedings were searched for relevant reports. Date of the most recent search: 2000. RCTs that compared one form of nutrition support with another, or control, for bone marrow transplant patients were included. Thirty five reports were identified, 11 were excluded. Two reviewers extracted data from 24 studies; 16 were allocated to four interventions: oral glutamine versus placebo; PN and glutamine versus standard PN; PN versus IV hydration; PN versus EN. Eight studies were other interventions. Data were collected on participants' characteristics; adverse effects; neutropaenia; % change in body weight; graft versus host disease; and survival. Trialists were contacted for unreported data. Two studies (82 subjects) found that glutamine mouthwash reduced days of neutropaenia (6.82 days, 95% CI (1.67-11.98) p=0.009) compared with placebo. Three studies (103 subjects) showed that patients receiving PN with glutamine had a reduced hospital stay, 6.62 d (95% CI 3.47, 9.77, P=0.00004) compared with patients receiving standard PN. Two studies (73 subjects) indicated that patients receiving PN plus glutamine had a reduced incidence of positive blood cultures (OR 0.23, 95% CI 0.08-0.65, p=0.006) compared to those receiving standard PN. One study, (25 subjects) showed patients receiving PN had a higher incidence of line infections (odds ratio 21.23, 95% CI 4.15,108.73, P=0.0002) compared to those receiving standard IV fluids. There were no evaluable data to compare PN with EN. Lack of evaluable data means that the relative effectiveness of EN versus PN cannot be evaluated. Further studies and missing data from completed trials need to be retrieved. Studies comparing PN with glutamine versus standard PN suggest that patients leave hospital earlier, and experience a reduced incidence of positive blood cultures, than those receiving standard PN. Patients with gastrointestinal failure should consider PN with the addition of glutamine if enteral feeding is not possible.