Dehydration in terminally ill patients is a common condition and its treatment in an active oncology unit often involves intravenous (IV) hydration programs. The decision to use IV hydration is only partly determined by an objective assessment of the clinical data. The other component is the attitudes of those involved in the decision-making process. This pilot study surveyed the attitudes of patients, family members/friends, nurses and doctors toward IV hydration in this patient population. Although the majority of the conscious patients (95%), family members/friends (81%), and nurses (64%) were not involved in the decision-making process, the expressed attitudes were generally positive. The three most commonly stated reasons for IV therapy were "medication," "giving fluids" and "giving morphine." Interestingly, among the reasons expressed, the amelioration of thirst was never explicitly cited.

Cancer patients in the terminal phase of their disease often experience fluid deficits. This is mainly due to their inability to ingest adequate amounts of oral fluids to meet the body's physiological demands. In order to correct this deficit, intravenous (IV) fluid programs are often instituted. This pilot study was conducted on a group of terminal patients hospitalized in an oncology unit who died while receiving IV fluids. It sought to assess the effects of these fluids on their level of thirst. Data were collected on 30 patients in the last 24 hours of life. However, of the 30 patients only 19 were sufficiently alert to be able to verbally evaluate their thirst intensity. Of the 19 patients, six experienced mild thirst, eight moderate thirst, and four severe thirst. This was in spite of IV hydration regimens which ranged from 500 mL to 3000 mL. Little relationship was found between level of thirst and the amount of IV fluids received, blood urea nitrogen (BUN), or sodium blood levels. In addition, although 70% of the patients had fluid retention signs, there was little correlation between these signs and the amount of fluids received. Since the pilot study's sample was small, definitive conclusions could not be drawn. However, our results highlight the need for future research in this area.

After observing what appeared to be the frequent misuse of intravenous (IV) hydration in terminally ill patients, we conducted a small research study in an active oncology department to determine the effects of IV hydration on signs of fluid retention in terminally ill patients. Dyspnoea 1.2 and edemas have been reported as two common symptoms experienced by patients with advanced cancer; shortness of breath and swollen legs were ranked 9 and 10 respectively in a list of distressing symptoms experienced by these patients.3 Although there are several reasons why terminal patients may experience dyspnoea and dependent edema, fluid overload may be one of the causes. Zerwekh suggested that administration of IV fluids increases the incidence of pulmonary secretions, peripheral edema and ascites.4We therefore chose to assess the presence and absence of rales, leg edema and ascites in those receiving IV hydration. We also wanted to observe any relationship between IV fluid volume, fluid output and these clinical signs.Data were collected on 21 patients who were receiving IV hydration on the day of death to determine the presence of rales, ascites and swollen legs. We found evidence of rales in 81%, ascites in 62% and leg edema in 52%. All except one patient showed evidence of one or more of these clinical signs (one sign 28%; two signs 28%; three signs 38%). However, little relationship was found between the incidence of signs and the net fluid balance (see table below). Two relationships did emerge, although, in view of the small numbers, neither reached statistical significance: patients with a history of hypertension or cardiac disease had a higher incidence of clinical signs than those patients who did not (two signs -43% vs. 19%; three signs -43% vs. 35%). There was also a relationship between fluid retention signs and age. The group of patients who showed evidence of one clinical sign had a mean age of 40 years (range 18-72), while those with two clinical signs had a mean age of 56 years (range 37-67), and those with three clinical signs had a mean age of 61 years (range 36-75).Previous published literature on the prevalence of dyspnoea and swollen legs among the terminally ill did not record the percentage of patients that were receiving IV fluids.'-' In one study on the use of hypodermoclysis (HDC) for the treatment of dehydration in terminal patients, the authors had only to discontinue HDC on three patients due to increased generalised edema and only one of their patients suffered from pulmonary edema 5 The differences between this study and ours may have been due to the difference in the method of fluid delivery. Further research needs to be conducted, into the management of dehydration in these patients, including a comparison of the effect of IV fluids, hypodermoclysis and no parenteral fluids on the incidence of signs of fluid retention.