Factors prompting sneezing in intravenously sedated patients receiving local anesthetic injections to the eyelids.

PURPOSE: To investigate the frequency of sneezing among patients receiving intravenous sedation and periocular local anesthetic for oculoplastic procedures in a single center. To identify potential risk factors involved.

DESIGN

Prospective, consecutive, interventional case series in a single tertiary-referral oculoplastic unit.

PARTICIPANTS

A total of 294 patients undergoing 314 isolated oculoplastic procedures, performed under intravenous sedation with periocular local anesthetic from November 2007 to November 2008.

METHODS

Prospective data collection on patient demographics, history of photic sneezing, intravenous sedative, depth of sedation, nasal oxygen, and periocular infiltration site. Standard local anesthetic was used in all cases, but the intravenous sedation was at the discretion of the attending anesthesiologist (7 in total).

MAIN OUTCOME MEASURES: Sneezing or attempted sneezing within 5 minutes of injection of the local anesthetic, as determined by agreed observation between attending staff.

RESULTS

Sneezing was observed in 16% of cases. No association was found between sneezing and patient age or presence of nasal oxygen. A weakly positive association was observed with male gender (55% sneezers vs. 37% non-sneezers, P = 0.03, relative risk [RR] = 1.5, confidence interval [CI], 1.1-2.0), bilateral infiltration (65% vs. 40%, P = 0.005, RR = 1.6, CI, 1.2-2.1), and upper eyelid infiltration (73% vs. 54%, P = 0.01, RR = 1.4, CI, 1.1-1.7). Photic sneezing was described in 47% of sneezers and 19% of non-sneezers (P = 0.0004, RR = 2.6, CI, 1.6-4.0). Because propofol was given to 95% of patients, no association with sneezing could be ascertained. However, opioid derivatives were found to be protective (12% vs. 43%, P<0.0001, RR = 0.3, CI, 0.1-0.6), whereas midazolam doubled the risk of sneezing (45% vs. 22%, P = 0.0008, RR = 2.1, CI, 1.4-3.0). Deep sedation (Ramsay score 5-6) also strongly increased the sneeze risk (65% vs. 23%, P<0.0001, RR = 2.8, CI, 2.1-3.8).

CONCLUSIONS

Propofol-based intravenous sedation, in combination with periocular local anesthetic injections, induces sneezing in approximately one sixth of general oculoplastic cases. Male gender, a history of photic sneezing, bilateral or upper eyelid infiltration, deep sedation, and the concurrent administration of midazolam all increased the risk, whereas adjunctive opioid use reduced the risk.