Pseudo-pulmonary oedema.

Sir,

Intraoperative pulmonary oedema can occur due to cardiogenic or non-cardiogenic reasons. It is diagnosed on the basis of the constellation of clinical features, monitoring parameters and investigations. Frothy (sometimes blood-stained) secretions in endotracheal tube (ETT), the pathognomonic sign of pulmonary oedema and the first sign to arouse suspicion, may have various aetiologies. We encountered a similar case in which search for sudden onset of frothy secretions from ETT led to diagnosis of iatrogenic cause for the condition.

A 56-year-old female with a history of chest pain was posted for elective coronary artery bypass grafting and had revealed no significant abnormality on pre-anaesthetic evaluation. As per protocol, ionotropes and vasodilators were prepared in infusion pumps. After applying routine invasive and non-invasive monitoring, patient was induced. Induction was uneventful and surgical procedure proceeded uneventfully. Intra-operatively end-tidal carbon-dioxide (ETCO2) monitor appeared to be malfunctioning, leading to improper capnogram. Since the partial pressure of CO2 was normal, and check of ETCO2 sensor could not be performed during the surgical procedure, it was done when patient was taken on cardiopulmonary bypass pump. Check was completed, and blocked gas sampling line was changed. After grafts were placed, weaning from bypass pump was initiated, nitroglycerine (NTG) infusion rate via infusion pump escalated as per routine practice and mechanical ventilation was resumed. Since no capnogram was evident, assuming side stream ETCO2 sensor failure, mainstream ETCO2 sensor was also placed in the circuit that showed normal capnogram and ETCO2 levels. During the surgical closure, we noticed frothy secretions in ETT. Immediately endotracheal and oral suction was done. O2 saturation was nearly constant, and no significant changes were noted in peak pressures while patient was mechanically ventilated. Conducted sounds were noted on auscultation that disappeared on suction. Endotracheal secretions were frequently accumulating which led to suspicion of pulmonary oedema. At the same time escalating doses of NTG were not effective in producing desired haemodynamic changes. Attempts to look for the cause revealed common aetiology to both the problems;– accidental misconnection of the distal end of drug infusing line of NTG infusion pump to side stream ETCO2 sampling port [Figure 1]. Hence, we termed it as pseudo-pulmonary oedema. This all happened during ETCO2 sampling line change. Similar diameter and connection port (luer lock ends) of the two lines, drug infusion line and gas sampling line, led to this misconnection [Figure 2]. Simultaneous use of multiple infusion pumps and their connection to the central line through three-way connectors along with intermingled cables of monitoring lines in limited space lead to delayed detection of misconnection. Since NTG infusion was prepared with saline under aseptic condition and the periodic suction was done from ETT no respiratory complication was noted in the post-operative period. Luer lock connectors have been reported to enable functionally dissimilar tubes or catheters to be connected resulting in serious injury or death to the patients.[123] However, we could not find literature for similar misconnection being reported earlier; hence the incident reported here seems first of its kind and thus rare.

Figure 1

The misconnection: Involved line marked with paper adhesive along its course

Figure 2

End-tidal carbon-dioxide sampling line similar in appearance to drug infusion line

This case highlights two useful aspects. First importance of proper labelling of drug infusion lines at patient end and tracing all lines back to their origin before making connections to avoid misconnections among themselves and with the sampling line of side stream ETCO2 monitor. Although this may take extra time, it's a necessary measure to prevent mishaps. Second, equipment design solutions – which either prevent the operator from making a misconnection or prompt him or her to make the connection correctly – such as physical incompatibilities between connectors, connectors with locking mechanisms and connectors with a distinct physical appearance are not effective by themselves because they rely on the user to both recognise them and use them correctly. They should be used in combination with other equipment design solutions and safe work practices.