Three patients were at the same hospital over the course of a few months for vascular access device (VAD) placement and experienced adverse outcomes. The adverse outcomes of two of them were secondary to drugs given for sedation, while the third patient’s situation was somewhat different.
A 31-year-old woman with end stage renal disease (ESRD) and pulmonary hypertension presented in the Vascular Access Unit for revision of a malfunctioning right internal jugular tunneled dialysis catheter. Her home medications were reviewed by a nurse and it was documented that she had been taking hydrocodone/acetaminophen 5/325 mg as needed for pain. She was not noted to be on any benzodiazepines or other opioids at the time of admission. The patient reported anxiety in the pre-operative area and received 1 mg of lorazepam followed by 50 mg of diphenhydramine intravenously. The patient then became very sleepy and was unable to maintain her airway; insertion of a nasopharyngeal airway was therefore required to continue the procedure. Her vital signs were stable throughout the procedure and no additional sedation was necessary. After the procedure, she remained very sleepy but arousable. After several hours of observation, her condition was considered to be stable and she was discharged. The day after the procedure, the patient was found at home in cardiac arrest and was pronounced dead at a different community hospital. Her toxicology report in the hospital emergency department was positive for amphetamines, opiates and benzodiazepines.
A 52-year-old man with a history of ESRD, heart failure, human immunodeficiency virus (HIV), and venous thromboembolism was admitted for left arm swelling and pain. The patient was taken to the Vascular Access Unit for a fistulogram and was given 100 mcg fentanyl and 2 mg midazolam intravenously for sedation. The patient became apneic and was unable to maintain his airway, so an oral airway was inserted, and naloxone 0.2 mg was given for reversal. The patient became arousable and able to maintain his airway without further incident.
A 47-year-old woman with ESRD, cardiac sarcoidosis, and type 2 diabetes mellitus presented in the Vascular Access Unit for outpatient percutaneous thrombectomy for a clotted fistula. Procedural sedation with 175 mcg fentanyl and 3.5 mg midazolam was given intravenously and the procedure was successful. Afterward, the patient became unresponsive due to a cardiac arrest during transport and a Code Blue was called. She was then transferred to the emergency department where she experienced another cardiorespiratory arrest and expired.
By Sierra Rayne Young, Pharm.D. and Iris Chen, Pharm.D., BCPS, BCOP
Vascular access procedures are extremely common and are beneficial for patients requiring intravenous (IV) access for more than seven to ten days.1 Patients with ESRD who receive hemodialysis must have functional vascular access so that large amounts of blood can flow continuously through the dialysis machine during treatments three to seven times each week.2 All three types of vascular access for hemodialysis — arteriovenous fistulae, arteriovenous grafts, and central venous catheters — can cause problems that require medical interventions, such as those described in the three cases above.3,4 These procedures are relatively short but may require the use of procedural sedation, which is usually very well tolerated but can involve significant risk, as these cases illustrate.
In the first case, both diphenhydramine and lorazepam were given within a three-minute time period. The higher end of the normal range of doses of these medications was administered without appropriately trialing lower doses first. Despite this fact, the doses administered were still considered reasonable for the procedure involved. Benzodiazepines were not on the patient’s home medication list, but she was known to have taken them on and off in the past. Lorazepam was likely selected for this patient because it had been utilized to treat anxiety during prior hospitalizations and the 1 mg dose of lorazepam had been previously tolerated.
Most vascular access procedural order sets also include midazolam, which may represent a better alternative to lorazepam. Midazolam has a short duration of action of 5,6 Short-acting opioids such as fentanyl and midazolam are commonly used together for conscious sedation,7 a sleepy-yet-arousable state balancing comfort and consciousness.1 Despite the safety of conscious sedation, some patients have an increased risk for adverse effects due to underlying comorbid conditions such as renal dysfunction, advanced age, complicated baseline respiratory status, and drug interactions.8 Given this patient’s poor renal function, midazolam may have been a more appropriate sedative for this short procedure.
After careful review by the hospital’s patient safety committee, it was determined that the death was not directly related to the medications administered during the patient’s procedure given that her vital signs and mental status were stable at discharge. Although there may have been some residual effect of the lorazepam and diphenhydramine, the patient apparently consumed illicit substances after discharge, based on the urine toxicology results. Although oversedation during the vascular access procedure was not likely the cause of death, this case still highlights the sensitivity of patients with ESRD to standard doses of sedating medications.
The second case involved a 55-year-old man with a notable history of ESRD and superior vena cava (SVC) thrombosis presenting for a fistulogram. During the procedure, he received fentanyl and midazolam, both within appropriate dosing limits. However, the case was complicated by the patient's compromised respiration, suggestive of superior vena cava syndrome (SVCS) and consistent with the patient’s past medical history. This syndrome can occur when there is an obstruction of blood flow through the SVC, potentially caused by malignancy, infection, aneurysm, or thrombosis.9 These patients face increased risk for respiratory compromise due to laryngeal edema and impingement of the airway.10 In this case, the clinical team did not identify the patient as at high risk for respiratory compromise despite his extensive SVC thrombosis. In the absence of a standardized process for screening patients preoperatively, high-risk patients such as these might not have appropriate staff present during their procedures to help preemptively control their airways. Other conditions that may place patients at high risk for oversedation and respiratory compromise during conscious sedation include chronic obstructive pulmonary disease, sleep apnea, asthma, and obesity.11
In this case, the health system’s own root cause analysis found that there was no process in place for collaborative, adjudicated decision-making between vascular access staff and anesthesia staff for patients deemed at high risk for complications. As a result of these first two cases, system-wide changes were implemented that included standardized pre-procedure screening by both a nurse and attending physician as well as electronic health record (EHR) enhancements with more decision support and flags for high-risk patients. Additionally, vascular access nurses and physicians launched a monthly Morbidity & Mortality meeting to discuss difficult cases and identify additional areas for improvement.
The third case involved a 48-year-old woman presenting for an outpatient percutaneous thrombectomy procedure. The patient tolerated the procedure without apparent issues and appropriate doses of fentanyl and midazolam were administered. After the procedure, she experienced cardiac arrest with return of spontaneous circulation upon moving to the gurney and was promptly brought into the emergency department where she later expired. Ultimately, the cause of death was determined to be consistent with an acute pulmonary embolism (PE). Percutaneous thrombectomy procedures are usually well tolerated and have a ninety percent success rate although there are certain factors, including infected fistulas, enlarged aneurysmal fistulas with extensive clot burden, repeated thrombosis, and hemodynamic instability, that increase the risk associated with this intervention.12 The patient in this case had an aneurysmal fistula with large clot burden, which has been associated with pulmonary embolism in up to 59 percent of patients.13 Fortunately, most of these PEs are asymptomatic, but the risk of symptomatic pulmonary embolism increases with larger clot burden.14
Although this case did not involve oversedation, it highlights other complications that can occur during vascular access procedures, especially among patients with ESRD. Better preoperative screening would have identified this patient as high-risk for complications given her aneurysmal fistula. The actions performed by the vascular access staff were appropriate and the patient’s death could not have been predicted. However, the acute onset of massive pulmonary embolism in the procedural setting is a reminder of the need for careful, frequent monitoring of all patients’ vital signs and clinical examinations during and after these procedures so that any problems can be identified and responded to as soon as possible.
Approach to Improving Patient Safety
While vascular access procedures remain extremely common and are usually successful, they can result in major complications. The cases presented above describe situations in which additional monitoring and staff could have been helpful for ensuring patient safety. Opioids and benzodiazepines remain agents of choice for procedural sedation; however, for patients with comorbid conditions affecting their respiratory status and renal dysfunction, special care should be taken. When procedural sedation is required, short-acting agents such as fentanyl and midazolam are preferred to avoid lingering adverse effects. Additionally, standardized pre-procedure screening by both nurses and physicians should provide more information about the patient’s past medical history, opioid or benzodiazepine tolerance, and any polypharmacy that may not be obvious from their current medication list. Clinical decision support features in the EHR can also help to alert medical staff to high-risk patients. Patients who are identified as high risk for procedural complications should have anesthesia staff present throughout the procedure to help manage the airway preemptively. Morbidity & Mortality meetings can also be helpful to discuss difficult cases and identify additional areas for improvement.
Vascular access procedures are extremely common and are typically well tolerated; however, major complications can include infections, embolism, bleeding, and over-sedation.
Opioids and benzodiazepines are commonly used for conscious sedation during vascular access procedures but should be used cautiously in patients with comorbid conditions such as renal dysfunction and respiratory compromise.
Patients should be thoroughly screened for risk factors that may lead to complications and, if any are identified, proper preventive measures—which can include contacting anesthesia and adjusting medications—should be taken.
Electronic health record (EHR) enhancements that provide decision support and flags for high-risk patients can increase the safety of procedural sedation.
Risk factors, including prior opioid/benzodiazepine usage and comorbid conditions, should be clearly identified and discussed in a huddle prior to the procedure.
The use of short-acting sedatives to prevent prolonged or oversedation should be encouraged, especially among patients with ESRD.
Sedating medications should be ordered and administered at the lowest doses needed to ensure that the patient with ESRD can tolerate the vascular access procedure.
Sierra Young, Pharm.D.
Clinical Pharmacist, Adult Critical Care
UC Davis Medical Center
Iris Chen, Pharm.D., BCPS
UC Davis Medical Center
1. Cheung E, Baerlocher MO, Asch M, Myers A. Venous access: a practical review for 2009. Canadian Family Physician. 2009 May 1;55(5):494-6.
2. Santoro, D., Benedetto, F., Mondello, P., Pipitò, N., Barillà, D., Spinelli, F., ... & Buemi, M. (2014). Vascular access for hemodialysis: current perspectives. International journal of nephrology and renovascular disease, 7, 281. [go to PubMed]
3. Rupp SM, Apfelbaum JL, Blitt C, Caplan RA, Connis RT, Domino KB, Fleisher LA, Grant S, Mark JB, Morray JP, Nickinovich DG. Practice guidelines for central venous access: a report by the American Society of Anesthesiologists Task Force on Central Venous Access. Anesthesiology. 2012 Mar;116(3):539. [go to PubMed]
4. Ruesch S, Walder B, Tramèr MR. Complications of central venous catheters: internal jugular versus subclavian access—a systematic review. Critical care medicine. 2002 Feb 1;30(2):454-60. [go to PubMed]
5. Fragen RJ. Pharmacokinetics and pharmacodynamics of midazolam given via continuous intravenous infusion in intensive care units. Clinical therapeutics. 1997 May 1;19(3):405-19. [go to PubMed]
6. McDermott CA, Kowalczyk AL, Schnitzler ER, Mangurten HH, Rodvold KA, Metrick S. Pharmacokinetics of lorazepam in critically ill neonates with seizures. The Journal of pediatrics. 1992 Mar 1;120(3):479-83. [go to PubMed]
7. Tobias JD, Leder M. Procedural sedation: a review of sedative agents, monitoring, and management of complications. Saudi journal of anaesthesia. 2011 Oct;5(4):395.[go to PubMed]
8. Ekström MP, Bornefalk-Hermansson A, Abernethy AP, Currow DC. Safety of benzodiazepines and opioids in very severe respiratory disease: national prospective study. Bmj. 2014 Jan 30;348:g445. [go to PubMed]
9. Cohen R, Mena D, Carbajal-Mendoza R, Matos N, Karki N. Superior vena cava syndrome: A medical emergency?. International Journal of Angiology. 2008 Mar;17(01):43-6. [go to PubMed]
10. Straka C, Ying J, Kong FM, Willey CD, Kaminski J, Kim DN. Review of evolving etiologies, implications and treatment strategies for the superior vena cava syndrome. Springerplus. 2016 Dec 1;5(1):229. [go to PubMed]
11. Zammit C, Liddicoat H, Moonsie I, Makker H. Obesity and respiratory diseases. International journal of general medicine. 2010;3:335. [go to PubMed]
12. Valentin CN, Zangan SM. Axillary vein rupture after angioplasty. InSeminars in interventional radiology 2009 Sep (Vol. 26, No. 03, pp. 276-278). © Thieme Medical Publishers. [go to PubMed]
13. Boonsrirat U, Hongsakul K. Pharmacomechanical thrombolysis for the treatment of thrombosed native arteriovenous fistula: a single-center experience. Polish journal of radiology. 2014;79:363. [go to PubMed]
14. Turetz M, Sideris AT, Friedman OA, Triphathi N, Horowitz JM. Epidemiology, pathophysiology, and natural history of pulmonary embolism. InSeminars in interventional radiology 2018 Jun (Vol. 35, No. 02, pp. 92-98). Thieme Medical Publishers. [go to PubMed]