A retrospective analysis of severe intraoperative respiratory compliance changes during ophthalmic arterial chemosurgery for retinoblastoma.

BACKGROUND AND OBJECTIVES

Ophthalmic arterial chemosurgery for retinoblastoma has been associated with intraoperative decreases in respiratory compliance. Through the analysis of data from computerized records, we objectively defined severe respiratory compliance events and correlated them with demographic and clinical information in patients undergoing this procedure.

METHODS

Data were collected from ophthalmic arterial chemosurgery cases from 2006 to 2013. Intraoperative PIP, PEEP, TV, SpO2 , and EtCO2 were analyzed. Compliance changes, desaturations, decreases in EtCO2 , and clinical outcomes were assessed.

RESULTS

Respiratory compliance decreases with a bimodal distribution. Severe events were defined as exhibiting a minimum compliance decrease of 40%. Seventy-eight of 122 children (64%) experienced a severe compliance event during at least one treatment, and it occurred in 137/468 cases (29%). A subset of 94 children had complete or at least the first three records. The incidence of a severe respiratory compliance event in this subgroup was 17/94 (18%) on the first and 84/261 (32%) on subsequent procedures. The probability of developing a severe respiratory compliance event on a subsequent procedure was 0.40 if the child developed it on the first procedure, 0.30 if he did not; this difference was not significant. The incidence of desaturation below 90% with severe respiratory compliance events was 0.20; the incidence of a 30% drop in EtCO2 was 0.34. No morbidity, no extended recovery, and no admissions were associated with intraoperative severe respiratory compliance events. We found no correlation between history, age, sex, weight or allergies, and intraoperative severe respiratory compliance events.

CONCLUSIONS

Here, most patients experienced a severe respiratory compliance event during at least one of their procedures. Overall incidence was 29% and was more likely on subsequent procedures. A severe respiratory compliance event at the initial procedure was poorly predictive of its occurrence on subsequent procedures. No morbidity was associated with intraoperative severe respiratory compliance events.