Clinical indicators of dehydration severity in elderly patients.

STUDY OBJECTIVE: To determine which of the signs and symptoms of dehydration obtainable from patient history and physical examination in the emergency department are most useful in assessing the severity of dehydration in elderly patients. DESIGN: Prospective, correlational study. SETTING: Two university teaching hospitals. PATIENTS: Fifty-five patients aged 60 or older presenting to the emergency department with suspected dehydration were studied. MEASUREMENTS AND MAIN RESULTS: In the emergency department, patients were evaluated by a standardized history and physical examination that included assessment of 38 signs and symptoms commonly attributed to dehydration. The relationships between the presence and intensity of these putative dehydration indicators and an independent rating of dehydration severity based on a comprehensive review of the medical record were evaluated. Also evaluated were the relationships between these dehydration indicators and patient age. Indicators that correlated best with dehydration severity but were unrelated to patient age included: tongue dryness (P less than 0.001), longitudinal tongue furrows (P less than 0.001), dryness of the mucous membranes of the mouth (P less than 0.001), upper body muscle weakness (P less than 0.001), confusion (P less than 0.001), speech difficulty (P less than 0.01), and sunkenness of eyes (P less than 0.01). Other indicators had only weak associations with dehydration severity or were also related to age. Patient thirst was unrelated to dehydration severity. CONCLUSIONS: A set of signs and symptoms related to dehydration severity in elderly patients has been identified. These indicators may be more useful for evaluation of dehydration severity in the emergency department than other commonly used indicators.

Deciding to restrain medical patients.

To determine when and why patients are placed in mechanical restraints, we surveyed the nurse and physician caring for each of 102 restrained patients from the general medical floors of an acute-care hospital. Ninety-three percent of the questionnaires were completed. Nursing questionnaires indicated that over half of patients were restrained during the evening shift. Nurses initiated the use of restraints in 75% of cases. Fifteen percent of the patients' physicians were unaware that the patient had been restrained. In the majority of cases, the nurse and physician believed that restraint was the best alternative for managing the patient although more physicians (11%) than nurses (2%) thought an alternative intervention would be better (P less than .02). As a group, physicians and nurses restrained patients for similar reasons, most often to prevent falls from bed (69%) or to protect medical devices (36%). However, there was poor agreement between the nurse and physician as to the reason for restraint in an individual patient (kappa statistic range from .02 to .43). These findings suggest that nurse and physician communication regarding restraint is poor. We recommend that acute-care hospitals adopt policies to promote communication between nurses and physicians concerning restraints to ensure that use of this potentially hazardous intervention is used only when necessary.

Mechanical restraints on the medical wards: are protective devices safe?

We prospectively identified 102 mechanically restrained hospital patients and determined their hospital course. The cohort was elderly, cognitively impaired with multiple chronic diseases. The in-hospital mortality was 21 percent. Nosocomial infection developed in 12 percent and new pressure sores in 22 percent. Prolonged use of restraints (greater than 4 days) was the strongest independent predictor of nosocomial infection (relative risk 1.8, 95% CI = 1.2, 2.8) and new pressure sores (RR 1.4, 95% CI = 1.1, 1.8) as determined by multiple logistic regression analysis. Patients placed in mechanical restraints for longer than four days experience frequent morbid events and should be monitored carefully.

Estrogen-progesterone therapy for recurrent gastrointestinal bleeding secondary to gastrointestinal angiodysplasia.

Recurrent gastrointestinal bleeding ceased after the institution of estrogen-progesterone therapy in a patient with gastrointestinal angiodysplasia without chronic renal failure. The patient did have aortic stenosis and mitral insufficiency murmurs. A therapeutic trial of conjugated estrogen should be considered in patients with recurrent gastrointestinal bleeding secondary to gastrointestinal angiodysplasia after the use of therapeutic endoscopy and/or surgery.

PIP: A case report is presented of a 78-year-old patient with recurrent gastrointestinal bleeding who required frequent hospitalizations and several blood transfusions, both of which ceased after the institution of estrogen-progesterone therapy. The patient was hospitalized in April 1985 with weakness and shortness of breath. Her hemoglobin was 9.8. Indices were suggestive of a microcytic hypochromic anemia. Bone marrow revealed absent iron stores. The patient denied a prior history of melena, hematochezia, or hematemesis. There was no history of peptic ulcer disease. The patient was taking no medications, and she denied alcohol abuse and a family history of bleeding. The physical examination revealed a pale and weak appearing women in no acute distress. An aortic stenosis and mitral insufficiency murmur was her only positive physical finding. Hemoccult positive stool was detected on rectal examination. The patient was transfused with 2 units of packed cells and underwent an evaluation. Subsequently, the patient required 4 more hospitalizations requiring blood transfusions. In October 1985, the patient underwent an exploratory laparotomy and an intraoperative endoscopy of the entire gastrointestinal tract. There was evidence of bleeding in the cecum near the ileocecal valve. The patient underwent an ascending colectomy. The histopathology of the specimen revealed angiodysplasia of the cecum. Postoperatively, during a 5-month period, the patient required 16 units of blood. In March 1985, Enovid E was started, and the patient has not required further hospitalizations or a blood transfusion since. The only side effects of the estrogen-progesterone combination which this patient experienced were vaginal bleeding and mild systolic hypertension. In patients with gastrointestinal angiodysplasia who have cryptogenic gastrointestinal bleeding, despite the use of therapeutic endoscopy and surgery, possibly a trial of estrogen-progesterone therapy should be considered.