Vascular Surgery

Case summary

A previously fit 37-year-old woman presented to a regional hospital. She was diagnosed with a left leg iliofemoral deep vein thrombosis (DVT) and associated pulmonary embolus. Exonaparin was started.

The patient was transferred to the local tertiary centre and admitted under Vascular Surgery, where a diagnosis of DVT associated with May-Thurner syndrome was presumed. Thrombolysis with an inferior vena cava (IVC) filter was planned. Three days after presentation the patient underwent the procedure under the care of the interventional radiologist. A filter was placed via the right groin under general anaesthetic in the angiography suite in the radiology department. Access for thrombolysis was from the left popliteal artery, where thrombolysis was achieved with urokinase prior to placing a stent (16 mm x 60 mm) in the left iliac vein. At or near completion of the procedure the patient became unstable and arrested. Cardiopulmonary resuscitation (CPR) commenced, and the patient was placed on veno-arterial extracorporeal membrane oxygenation (ECMO) and transferred to the intensive care unit. Intra-abdominal bleeding was identified via computed tomography (CT) scan with the stent in the appropriate position. An attempt at embolisation failed to identify an arterial source. Despite all efforts, aggressive care was futile, and the patient died very soon afterwards in the early hours of the day after surgery.

Discussion

This was a fit young woman who died as a consequence of intervention for a condition that was not life-or limb-threatening.

In the immediate period after the arrest, multiple vascular surgeons were involved in the patient’s care. Decisions about the direction of care seem to have been taken as a multidisciplinary team and the family appears to have been involved very closely. The assessor had no concerns about communication or that all resources were pooled to try and manage this case.

The likely cause of the initial arrest was massive pulmonary embolus, despite preoperative placement of an IVC filter. The severe cardiac failure and loss of oxygenation that followed was only partially resolved with ECMO. Whether or not the situation was salvageable at this point is uncertain, but massive intra-abdominal bleeding occurred, which was not survivable for this patient.

A clear gap in the records is a diagnosis for this massive intra-abdominal bleeding (15 units blood transfusion) progressing to abdominal compartment syndrome within a few hours. Failure to identify an arterial source may reflect hypotension and increased abdominal pressure, rather than the true absence of an arterial source. The initial operation was a venous procedure, so it is difficult to see how an arterial injury could have occurred. However, an arterial injury may have occurred during insertion of the ECMO cannulas. Another possibility is that the ilio-caval venous system was ruptured by instrumentation or stent/angioplasty, but this is a low-pressure system, and it is difficult to know how this would have caused abdominal compartment syndrome.

Vena cava rupture during CPR has also been described, and in the presence of right ventricular outflow obstruction may be more likely. Certainly, anticoagulant and thrombolytic use would have exacerbated the bleeding. This problem seems to have made ECMO significantly more difficult to maintain. While it is possible that the pulmonary embolisation was survivable with ECMO, in combination with abdominal compartment syndrome it was not.

At the time of review, the coroner’s postmortem findings were unavailable, so the cause of death is unclear. Thus, it is difficult to comment on what went wrong. The assessor could only comment to express significant concerns about the outcome of this case. It is equally possible that there was a surgical accident of one sort or another, or that this was just a very unfortunate series of events. This case was subject to extensive internal review. If there were indications of less than excellent practice in this case, this would have been addressed in a professionally appropriate manner.

In summary, this was a tragic case. A young person was subject to an invasive procedure for a condition that, at presentation, would have held a relatively low risk of mortality (though undeniable risk of long-term morbidity). The patient died as a consequence of a combination of a massive pulmonary embolus and intra-abdominal bleeding that suggests interventional misadventure. That thrombolysis with an IVC filter is increasingly considered a standard of care for this condition should not diminish our concern.

Clinical lessons

Invasive procedures may be attended by disaster, no matter how routine the procedure, how fit the patient is, or if the case appears to be uncomplicated. This should always be remembered when obtaining patient consent prior to intended procedures. Instrumentation of fresh thrombus in large veins carries risk of pulmonary embolus and death.