Too Many Cooks in the Kitchen
- Explain the role of shared decision-making and “stopping rules” in planning operative interventions in patients with multiple chronic comorbidities
- List at least two safety-related disadvantages of starting complex surgical cases late in the day
- Describe the problem of perioperative fixation error and approaches to help minimize the risk of such error
- Describe the role of decision-support alerts in postoperative care, and identify several factors that might trigger such alerts based on cardiac risk after non-cardiac surgery
A 40-year-old man with diabetes, hyperlipidemia, hypertension, coronary artery disease (status post percutaneous coronary intervention), chronic renal failure (on hemodialysis), and severe aortic stenosis (aortic valve area 0.64 cm2), was admitted to a major trauma center for a pathologic pelvic fracture (secondary to osteoporosis) after a fall from standing. The Nephrology, Cardiology, and General Internal Medicine services were involved, but, as per routine at this center, Trauma Surgery was the primary team. Orthopedics considered the fracture stable based on post-mobilization x-rays. The patient's recovery was slow due to ileus and severe deconditioning.
While being assisted to a standing position on hospital day 13, the patient’s left leg buckled and he developed hip pain. X-rays showed an intracapsular left femoral neck fracture with severe osteopenia secondary to renal osteodystrophy. Cardiology, Cardiothoracic Surgery, Trauma Surgery, and Orthopedics discussed this high-risk case extensively and ultimately planned operative hemiarthroplasty with a cardiac anesthesiologist, including intraoperative trans-esophageal echocardiography (TEE). The patient was not considered to be a candidate for preoperative open or trans-catheter aortic valve replacement or balloon valvuloplasty for his aortic stenosis. Due to limited operating room availability, his hip operation on hospital day 24 began at 16:00 and concluded at 22:30. The operation, performed with general anesthesia, was prolonged due to unsuccessful epidural catheter placement and an intraoperative periprosthetic fracture, requiring plating, but otherwise was uneventful. Neither a cardiac anesthesiologist nor TEE was involved. The patient was hypertensive during his 90-minute stay in the Post-Anesthesia Care Unit (PACU). He was then transferred back to a surgical unit (not intensive care) with only routine vital signs and nursing checks every four hours for hemodynamic monitoring. His initial systolic blood pressure after transfer was 60 mmHg. The on-call Trauma Surgery team was notified one hour later and administered 500 mL of intravenous fluids. En route to the Intensive Care Unit, the patient arrested and died.
By Richard P. Dutton, MD, MBA
This patient represents a challenging trauma admission. Based on a review of over 2,000 deaths in a major US trauma center, this patient had a high risk for mortality; trauma patients who die in the subacute phase typically have multiple comorbidities.1,2 While the injury itself (stable pelvic fracture) is easily survived in most cases, the evident frailty of this patient was a valid cause for concern.3,4 Severe aortic stenosis is a known risk for anesthesia,5 and the high relative mortality conferred by this condition extends to any other physiologic stressor, including recovery from trauma. The trauma surgery team and orthopedic consultant were correct to adopt a deliberate and conservative course; successful physical therapy may have been this patient’s best chance for meaningful recovery. It is not surprising that a trauma surgeon directed this patient’s care – coordination of a team of experts is an important skill, emphasized in trauma and emergency general surgery training – and the literature suggests this approach works well.6,7
Not surprisingly, however, this patient developed a serious complication. Ileus – possibly induced by use of opiate pain medications – prolonged the patient’s hospital stay and impaired mobilization. One of the most insidious iatrogenic insults affecting older patients with serious injuries is the inability to maintain muscle strength and mobility. Prolonged hospital bedrest leads to loss of muscle and bone mass and increases the risk of pressure ulcers, venous thromboembolism and hospital acquired-infection.8 Pneumonia leading to sepsis and organ failure is a leading cause of death in immobile older trauma patients.9
In this case, efforts at mobilization led to a second fracture, affecting the proximal femur. Repair of an osteoporotic fracture is harder than in healthy bone: the team must be prepared for substantial blood loss and the possibility of insufficient healthy bone to effect a repair. Indeed, the need to add a plate following fracture repair reflected this reality.
Approach to Improving Safety & Patient Safety Target
The most obvious management failure in this case, in retrospect, was the decision to proceed to a major surgical procedure. With the benefit of hindsight, it seems obvious the interests of the patient, the family and the healthcare system would have been better served by a palliative care consult and possible institution of comfort-care measures in the absence of any other realistically achievable goals. Was the eventual outcome predictable enough to justify this criticism? What were the odds of meaningful recovery if everything had gone well? Based on what is presented about the case, it seems more likely than not – but not certain – that the patient would have died before leaving the hospital, regardless of the approach. Lingering uncertainty leaves an important gap, which must be filled with robust discussion amongst the team and then shared decision making about the goals of care between the physicians, the patient and the family. Failure to have this discussion would be inexcusable. A further enhancement would have been detailed discussion of potential intraoperative events, with ‘stopping rules’ to limit increasingly futile efforts. For example, an agreement not to perform cardiopulmonary resuscitation (CPR) if cardiac arrest occurred, or a preplanned decision to terminate the procedure if transfusion was needed.10
It appears the team recognized the need for multidisciplinary discussion and developed a detailed plan for surgery. However, the plan was not effectively implemented. Whether this failure was due to inertia, a breakdown in communications, or inappropriate production pressure is not certain; likely all of these factors contributed to some degree. The case was allowed to proceed at 1600, rather than as a first case start in the morning. Morning starts are recommended for high risk cases when possible, both to allow proper marshalling of resources and to have all members of the team at their intellectual peak.11 Making this an add-on case at the end of the day also extends the risk for an adverse outcome beyond the index patient; performance of a complicated and resource consumptive operation in the evening distracts attention from the care of other patients in the operating suite, potentially at a time when personnel and intellectual resources are already stretched thin.
The procedure itself was complicated. An experienced orthopedic surgeon – and this case should have been performed by the most experienced one available – can complete a hemiarthroplasty in an hour of operative time, but this patient was in the room for more than 6 hours. Prolonged operative time is a surrogate for tissue injury, inflammatory load, fluid volume shifts, exposure to anesthetic medications, heat loss and infection risk.12 Proper care of this patient should have included reducing the physiologic impact of the surgery by shortening operative time, including aborting to a palliative procedure [e.g., percutaneous pinning of the femoral head, or even femoral head resection (Girdlestone procedure)] if indicated by the orthopedic findings. Failure to modify a surgical plan in the face of a difficult procedure is a common variant of perioperative fixation error, a cognitive error involving inordinate focus on one possible approach while ignoring others, akin to anchoring bias.13,14 Learning tools that teach “outside-the-box” or “lateral” thinking have been shown to circumvent fixation errors by increasing situational awareness. Other strategies for overcoming fixation errors including ruling out the worst-case scenario, accepting that the first assumption or original plan may be wrong, considering artifacts as the last explanation for a problem, and not biasing team members with any previous recommendations or conclusions.
Without a detailed review of the anesthesia record, it is difficult to comment on the specifics of management. The absence of a cardiac anesthesiologist is a minor concern, given that most anesthesiologists have plenty of experience managing elderly and frail patients for procedures such as this one, especially in Level 1 trauma centers. The failure to use TEE for intraoperative monitoring is more significant; while virtually all cardiac anesthesiologists are comfortable placing a TEE probe and interpreting images (perhaps the reason one was recommended in this case), this skill is less widespread in the general community. Use of TEE allows direct visualization of cardiac filling and contractility, as well as assessment of aortic valve function, right-heart distension and embolic load during manipulation of the fracture and placement of the femoral hardware. Use of TEE to guide ongoing resuscitation would have enabled management of the patient’s hemodynamics within a tighter range of optimal performance.
The fact that the patient survived to reach the recovery room and was then documented as hypertensive suggests that intraoperative fluid resuscitation was at least adequate, if not excessive. The team apparently had trouble placing an epidural catheter, which would have been strongly indicated in this case. Other options besides simply defaulting to general anesthesia include intrathecal or regional anesthetic approaches. In skilled hands, an epidural approach would have eased the sudden shifts in sympathetic tone associated with the onset and resolution of general anesthesia, mitigating the potential for abrupt fluid shifts. Sudden hypotension can cause rapid decompensation in patients with aortic stenosis. Epidural analgesia would have also helped avoid pain-induced tachycardia, which may well have been the proximate cause of the patient’s abrupt and terminal deterioration. Tachycardia causes a rise in myocardial oxygen demand relative to supply; so-called ‘demand ischemia’ is a well-recognized cause of perioperative myocardial infarction and is an increased risk in patients with aortic stenosis. Even in patients healthier than this one, a perioperative myocardial infarction may be associated with high short-term mortality typically quoted in the range of 5-25%.15 In this patient, the odds of successful resuscitation following cardiac arrest would be very low.
It is obvious that a major miscommunication occurred in the postoperative care of this patient. From the description of events, it appears the patient was managed in routine fashion when every aspect of the case should have dictated heightened attention. This reflects normalization of deviance in post-anesthetic care; because most patients recover uneventfully, there is a subconscious assumption that all patients will do so. This patient should have been transferred directly to an intensive care unit from the operating room, with continued advanced hemodynamic monitoring and close observation for any sign of distress. Both postoperative fluid shifts after a major surgery and the ongoing need for analgesia and heart rate control should have been anticipated. The ICU team should have been strongly warned that this patient was at high risk for abrupt decompensation and death.
Systems Change Needed/Quality Improvement Approach
Several system improvements might have prevented this event. The different teams of physicians should have followed through on their preoperative multidisciplinary discussion. All parties should have been automatically notified that the operation was occurring, so they could be prepared to see the patient again postoperatively in the ICU. Within the anesthesiology team, the use of a structured handoff tool could have improved retention of planning information between the day crew – that presumably consulted on the patient – and the on-call team that ended up doing the case.16
Within modern electronic medical records, it is possible to create decision-support alerts that identify patients at high risk for postoperative morbidity; such alerts can enable a hospital policy for mandatory ICU admission, or at least a consultative visit (e.g., by cardiology) in the PACU.17 Given that severe aortic stenosis has long been recognized as a major risk factor for perioperative cardiac events,18,19 this patient would have triggered such an alert, including mandatory transfer to the ICU for early postoperative care. A recent systematic review of 11 cardiac risk indices involving 2,910,297 adult patients who underwent noncardiac surgery found that the factors most consistently predictive of adverse cardiac outcomes were heart failure, type of surgery, serum creatinine, diabetes, history of stroke or transient ischemic attack, and emergency surgery.20 Substantial data from the largest studies also supported advancing age, American Society of Anesthesiology physical status classification, functional status, and hypertension as additional risks.
While most patients are discharged from PACU to a regular medical/surgical unit “per protocol,” a hospital policy requiring an in-person visit by the attending anesthesiologist or primary team physician prior to PACU discharge would help to protect high-risk patients. This policy could be triggered by patient characteristics (e.g., American Society of Anesthesiologists Physical Status 4 or 5), high-risk operations (e.g., carotid endarterectomy), or the occurrence of respiratory or hemodynamic instability in the operating room or PACU.
As a late-occurring safeguard that might have caught this patient’s deterioration before the outcome became inevitable, a hospital-wide Early Warning System based on networked bedside monitors could have triggered an early, automated call to the institutional rapid response team. Research has shown that such systems improve response time and overall patient outcomes.21
The patient’s death was not surprising and might have occurred even with optimal management. The failure of communication, however, both in deviation from a reasonable plan for management of anesthesia and in sending the patient to routine postoperative care rather than the ICU clearly contributed to his demise. This sad story reflects numerous opportunities for improvement in the care of a high-risk surgical patient, both in identification and communication of the risk and in the use of appropriate mitigating strategies.
Take Home Points
- Frail patients represent high-risk candidates for surgery and require multidisciplinary planning to achieve optimal results.
- Once the plan is developed for such a patient, it is important to follow it as prescribed. Deviation in the face of production pressure is a recipe for disaster.
- Optimal care of frail patients hinges on a plan with the least possible perioperative stress, both surgical and anesthetic. The team should strive for shorter surgical time, less tissue trauma and reduced need for fluid resuscitation or transfusion.
- For frail patients undergoing high-risk procedures, postoperative high-intensity care is required to avoid decompensation, or to detect and respond rapidly to early signs of decompensation.
Richard P. Dutton, MD MBA
Chief Quality Officer, US Anesthesia Partners
Adjunct Professor, Department of Anesthesiology
Texas A&M University College of Medicine
- Dutton RP, Stansbury LG, Leone S, Kramer E, Hess JR, Scalea TM. Trauma Mortality in Mature Trauma Systems: Are We Doing Better? An Analysis of Trauma Mortality Patterns, 1997-2008. J Trauma 69(3):620-6, 2010. PMID: 20093983
- Osler TM, Glance LG, Cook A, Buzas JS, Hosmer DW. A trauma mortality prediction model based on the ICD-10-CM lexicon: TMPM-ICD10. J Trauma Acute Care Surg. 2019;86(5):891-895. doi:10.1097/TA.0000000000002194
- McIsaac DI, MacDonald DB, Aucoin SD. Frailty for perioperative clinicians: a narrative review. Anesth Analg: 2020; 130: 1450-60
- Poulton A, Shaw JF, Nguyen F, Wong C, Lampron J, Tran A, Lalu MM, McIsaac DI. The association of frailty with adverse outcomes after multisystem trauma: a systematic review and meta-analysis. Anesth Analg: 2020; 130: 1482-92
- Kwok CS, Bagur R, Rashid M, et al. Aortic stenosis and non-cardiac surgery: A systematic review and meta-analysis. Int J Cardiol. 2017;240:145‐153. doi:10.1016/j.ijcard.2017.04.037
- Mathur S, Lim WW, Goo TT. Emergency general surgery and trauma: Outcomes from the first consultant-led service in Singapore. Injury. 2018;49(1):130‐134. doi:10.1016/j.injury.2017.09.002
- Mackenzie EJ, Rivara FP, Jurkovich GJ, et al. The National Study on Costs and Outcomes of Trauma. J Trauma. 2007;63(6 Suppl):S54-S86. doi:10.1097/TA.0b013e31815acb09
- Truong B, Grigson E, Patel M, Liu X. Pressure Ulcer Prevention in the Hospital Setting Using Silicone Foam Dressings. Cureus. 2016;8(8):e730. Published 2016 Aug 8. doi:10.7759/cureus.730
- Salarbaks AM, Lindeboom R, Nijmeijer W. Pneumonia in hospitalized elderly hip fracture patients: the effects on length of hospital-stay, in-hospital and thirty-day mortality and a search for potential predictors [published online ahead of print, 2020 May 21]. Injury. 2020;S0020-1383(20)30424-1. doi:10.1016/j.injury.2020.05.017
- Kopecky K, Pelletier P, Miller P. Strategies for Collaborative Consideration of Patients' Resuscitation Preferences. AMA J Ethics. 2020;22(4):E325-E332. Published 2020 Apr 1. doi:10.1001/amajethics.2020.325
- Kelz RR, Tran TT, Hosokawa P, et al. Time-of-day effects on surgical outcomes in the private sector: a retrospective cohort study. J Am Coll Surg. 2009;209(4):434‐445.e2. doi:10.1016/j.jamcollsurg.2009.05.022
- Turrentine FE, Wang H, Simpson VB, Jones RS. Surgical risk factors, morbidity, and mortality in elderly patients. J Am Coll Surg. 2006;203(6):865‐877. doi:10.1016/j.jamcollsurg.2006.08.026
- Fioratou E, Flin R, Glavin R. No simple fix for fixation errors: cognitive processes and their clinical applications. Anaesthesia. 2010;65:61–69. doi: 10.1111/j.1365-2044.2009.05994.x.
- Ortega R. Nasrullah K. On Reducing Fixation Errors. Anesthesia Patient Safety Foundation Newsletter. 2019;33(3). https://www.apsf.org/article/on-reducing-fixation-errors/
- Devereaux PJ, Xavier D, Pogue J, et al. Characteristics and Short-Term Prognosis of Periperative Myocardial Infarction in Patients Undergoing Noncardiac Surgery. A Cohort Study. Annals of Internal Medicine. 2011;154:523-8. doi: 10.7326/0003-4819-154-8-201104190-00003.
- Terekhov MA, Ehrenfeld JM, Dutton RP, Guillamondegui OD, Martin BJ, Wanderer JP. Intraoperative Care Transitions Are Not Associated with Postoperative Adverse Outcomes. Anesthesiology. 2016 Oct;125(4):690-9. PMID: 27466034
- Tremper KK, Mace JJ, Gombert JM, Tremper TT, Adams JF, Bagian JP. Design of a Novel Multifunction Decision Support Display for Anesthesia Care: AlertWatch® OR. BMC Anesthesiol. 2018;18(1):16. Published 2018 Feb 5. doi:10.1186/s12871-018-0478-8
- Goldman L, Caldera DL, Nussbaum SR, et al. Multifactorial index of cardiac risk in noncardiac surgical procedures. N Engl J Med. 1977;297(16):845-850. doi:10.1056/NEJM197710202971601
- Christ M, Sharkova Y, Geldner G, Maisch B. Preoperative and perioperative care for patients with suspected or established aortic stenosis facing noncardiac surgery. Chest. 2005;128(4):2944-2953. doi:10.1378/chest.128.4.2944
- Wright DE, Knuesel SJ, Nagarur A, Philpotts LL, Greenwald JL. Examining Risk: A Systematic Review of Perioperative Cardiac Risk Prediction Indices. Mayo Clin Proc. 2019;94(11):2277-2290. doi:10.1016/j.mayocp.2019.03.008
- Liu VX, Lu Y, Carey KA, et al. Comparison of Early Warning Scoring Systems for Hospitalized Patients With and Without Infection at Risk for In-Hospital Mortality and Transfer to the Intensive Care Unit. JAMA Netw Open. 2020;3(5):e205191. Published 2020 May 1. doi:10.1001/jamanetworkopen.2020.5191