Summary Factors associated with the failure of intravenous infusions due to phlebitis and extravasation were studied with 218 infusions delivering cytotoxic drugs. The survival rate of these infusions was not significantly different from that of 56 non-cytotoxic infusions in oncology patients. Although survival analysis indicated that cisplatin was associated with longer survival, this was probably an artifact caused by this drug usually being preceded by 24 h prehydration. Multivariate analysis indicated that etoposide was the only drug associated with decreased infusion survival and that bleomycin, cyclophosphamide, doxorubicin, ifosphamide, methotrexate, treosulphan and 5-fluorouracil had no significant effects. Also age of patient, infusion site and flow rate had no effects but survival was shorter in women. Follow-up When an infusion site could be located precisely at a subsequent visit (from notes and the patients memory), patency of the vein proximal to the infusion site was determined by inspection and palpation.
ResultsOf 284 infusion sites studied, 78 (27.4%) failed with 32 extravasating, 41 becoming phlebitic and five extravasating with obvious phlebitic signs. Of the failed infusions, 46 (13 for extravasation and 32 for phlebitis) required resiting for treatment (usually less than 72 h) to continue. In addition, four, three, one and one infusions were resited because of 'blockage', dislodgement, clotting and leakage respectively but these were not classed as failed.Statistics for oncology patients such as age, sex, cannula site and average fluid rate are shown in Table I. None of these factors was significant by univariate analysis.Cytotoxic drugs were infused at 228 sites while the 56 other sites received only crystaloid fluids and sometimes antibiotics. There was no significant difference between cytotoxic and non-cytotoxic infusions (RRFs 1.02 and 0.99 respectively, X2=0.01) (Figure 1).The most common drug given was cisplatin with bleomycin, etoposide, methotrexate, treosulphan and 5-fluorouracil (5-FU) also being used frequently. Other drugs (mitozan, JM8 and vinblastine) were used only occasionally (three, two and one times respectively). Univariate analyses of survival of infusions with individual drugs (Table II) showed that only cisplatin was associated with a statistically significant difference (this was longer survival). Table III shows factors selected by the multifactorial analysis. Cisplatin was associated with longer survival while infusions in females and infusions with etoposide had shorter survival.Subsequent patency was determined for 39 sites. Of 34 which ended without failure, 28 were patent while six were