Post‐tonsillectomy Bleeding as a Stochastic Process

Objective 1) Establish an acceptable post‐tonsillectomy bleeding rate from the literature, and 2)provide a literature‐based comparator for institutional bleed rates. Methods A MEDLINE literature search entering the keyword “tonsillectomy” identified 4,610 papers for all available years. The abstracts were loaded into a database and searched to identify the incidence of bleeding if reported. The weighted mean, standard deviation and 95% confidence intervals were calculated for those papers. For papers that compared a new technique to a control technique, only the control group rate was used. The definitions of bleeding rate were those used by the original authors of the papers. Results 63 papers reported post‐tonsillectomy bleeding rates. The mean (4.5%) plus 2 standard deviations (9.4%) suggests a maximum “expected” sustained bleeding rate of 13.9%. Even in this literature which should reflect optimum results, there were 3 reports of bleed rates in the 18–20% range. Conclusions Post‐tonsillectomy bleeding rates of about 5% are typical. Rates above 14% justify monitoring, and sustained rates over 20% occur at times. Transient increases greater than 20% may occur.