Gasembolie mit Kreislaufstillstand bei Hysteroskopien: Fallberichte von drei Patientinnen

 

PDF Download Buy Article Permissions and Reprints

Zusammenfassung

In einem Zeitraum von einem Jahr trat bei 3 von 62 Patientinnen ein Kreislaufzusammenbruch mit Asystolie während einer routinemäßigen Hysteroskopie auf. Alle konnten ohne neurologische Residuen reanimiert werden. Die wahrscheinlichste Ursache war in allen Fällen eine massive CO₂-Embolie. Sechs weitere Patientinnen wurden dann bei Hysteroskopien mit einem präkordialen Dopplerstethoskop bzw. einem Ösophagusstethoskop überwacht. In drei Fällen wurden typische Geräuschphänomene der Gasembolie nachgewiesen, ohne kardiovaskuläre Veränderungen. Die routinemäßige Überwachung des rechten Vorhofes und Ventrikels mit einem Doppier- bzw. Ösophagusstethoskop ist empfehlenswert, damit Gasbläschen frühzeitig erkannt und die Entwicklung potentiell letaler Komplikationen durch Drosselung der Gaszufuhr verhindert werden können. Auf Lachgas im Beatmungsgas sollte verzichtet werden, da N₂O die Resorptionszeit der CO₂-Bläschen verlängert. Alternativen sind die totale intravenöse Anästhesie, Regionalanästhesie bzw. Inhalationsanästhesie mit höheren Konzentrationen des volatilen Anästhetikums. PEEP-Beatmung als Prophylaxe ist aufgrund der Druckverhältnisse ohne Nutzen. Neben den allgemeinen Richtlinien der kardiopulmonalen Reanimation gehören zu den besonderen Maßnahmen zur Beherrschung embolischer Komplikationen die Kopftieflagerung in Linksseitenlage und die Hyperventilation mit reinem Sauerstoff.

Summary

Hysteroscopy and laparoscopy are relatively non-invasive methods routinely employed in the investigation of infertility. The danger of air embolism during these procedures was recognized early, and carbon dioxide was substituted for air since it is more readily soluble in blood. In this report we describe 3 cases of circulatory collapse and cardiac arrest in healthy young women during routine hysteroscopy (out of a total of 62 patients during the period 1989 - 1990) which were most probably caused by massive carbon dioxide embolism. Premedication was with oral diazepam 10 mg. Anesthesia was induced with 0.1 mg fentanyl, 2.5 mg droperidol and 100 mg methohexital (100 mg propofol in one case). Intubation was facilitated with 2mg pancuronium and 50 - 100 mg succinylcholine. Anesthesia was maintained with nitrous oxide 66 % and halothane. Ventilation was controlled with a tidal volume of 10 ml per kilogram body weight at a rate of 10 per minute. Monitoring included ECG, automated non-invasive blood pressure, capnometry, pulse oximetry and body temperature. Anesthesia was uneventful prior to insufflation. In each case the signs and symptoms began approximately 5 - 8 minutes after the start of insufflation and consisted of an initial tachycardia rapidly followed by ventricular dysrhythmias, bradycardia and cardiac arrest. The end-tidal CO₂ decreased during the tachycardic phase and prior to asystole. The patients were cyanotic with engorged jugular veins. Resuscitation with closed chest heart massage and intravenous epinephrine or orciprenaline was successful in every case. The typical "mill wheel phenomenon" of gas embolism was audible on auscultation after heart activity had returned, but disappeared after about 5 minutes. Two of the patients had a prolonged awakening with uncoordinated arm movements and lack of response to verbal commands or painful stimuli. They were given 40 mg dexamethasone, transferred to the intensive care unit and ventilated for several hours after which time they were able to be extubated. All 3 patients recovered with no permanent neurological sequelae. Alerted by these complications, we monitored right atrial and ventricular sounds with a precordial Doppler or esophageal stethoscopes in a series of six patients. We found that clinically inapparent gas embolism occurred in three of the patients; an incidence much higher than suspected. We therefore recommend the routine use of a Doppler stethoscope applied to the right sternal border or of an esophageal stethoscope during hysteroscopy and laparoscopy. This would allow early recognition of gas in the right heart and enable the anesthesiologist to warn the gynecologist of the impending danger and stop the gas flow. Positive end-expiratory ventilation is of little use in preventing embolism, since the intrauterine pressures reach values of 100 mmHg and more. Nitrous oxide should be omitted during anesthesia; it does not increase the size of the CO₂ bubbles as it does with air, but it has been shown to delay their resorption, thus allowing them to accumulate. The safety limit of 100 mg · min^-1 for the maximum CO₂ flow rate was determined in animals breathing atmospheric air. This rate must obviously be reduced considerably to allow for the delayed resorption of CO₂ bubbles in the presence of nitrous oxide. Alternatives would be total intravenous anesthesia, regional anesthesia or inhalational anesthesia with higher concentrations of the volatile anesthetic. In cases of clinically significant embolism the patients must be brought into a head-down, left-tilt position, and hyperventilated with 100 % oxygen to prevent hypoxia and facilitate CO₂ removal. Advanced cardiopulmonary resuscitative measures should be applied as necessary. Successful resuscitation is much more likely following embolism with CO₂ than with air due to its better solubility, although paradoxical embolism with neurological symptoms has been reported.