- Question: What are the significant perioperative adverse events reported in pediatric patients in a tertiary care center in a low- and middle-income country?
- Findings: Respiratory events were most common, and infants, especially those with congenital heart disease, formed a higher proportion of the group who had a perioperative cardiac arrest.
- Meaning: The decline in perioperative cardiac arrest events may have been due to changes in drugs and practice improvements.
Respiratory and cardiovascular adverse events in the perioperative period have been reported as major causative factors in anesthesia-related morbidity and mortality (M&M) in pediatric anesthesia.1 Analysis of these adverse events helps to identify subgroups of children who are at high risk. In addition, these audits help modify clinical practice, justify resource allocation, and direct future research projects.2
Two previous studies have reported on perioperative cardiac arrest in children from Pakistan and India,3,4 but there are limited epidemiological data available on serious adverse events occurring during the perioperative period from South Asian countries. Regional data are important since most South Asian countries are low or middle income and have similar issues related to the availability of staff, drugs, and equipment. In addition, these human and infrastructure deficits could lead to increased perioperative M&M in these regions.
Our aim was to assess and estimate the proportion of pediatric patients with various severe adverse events, following general or regional anesthesia in our tertiary care teaching hospital, over a 25-year period from 1992 to 2016. Our primary outcome was any intermediate and severe morbidity or mortality that occurred intraoperatively or within 48 hours postoperatively. Our secondary objective was to describe the clinical situations and causes of these events.
This study was approved by the institutional review board (IRB) of Aga Khan University. The requirement for the patient- or parental-written informed consent was waived by the IRB due to the nature of this review. We followed the departmental policy for access and retrieval of data.
Our hospital serves as a referral center for the southern region of the country. Between the years 1992 and 2005, a 25-year period, the hospital grew from 450 to 650 beds, and the number of operating rooms (ORs) increased from 8 to 16 rooms. Although there are some dedicated pediatric lists, pediatric patients have also been located within the lists of orthopedics, neurosurgery, plastics, ophthalmology, and ear, nose, and throat (ENT).
Anesthesia consultants assigned to pediatric lists undergo a credentialing process set up by the department, but they are not limited to the practice of pediatric anesthesia alone and work in other specialties as well. Only pediatric cardiac surgery has a dedicated group of consultant anesthesiologists and has a separate database of M&M.
All intermediate to major M&M details are maintained in a departmental database. All cases are reported, based on a mandatory requirement, to the departmental coordinator. These cases are then discussed in the monthly departmental M&M meeting. Three forms are generated for each case; a summary form filled by the primary anesthetist and 2 additional review forms filled by the 2 independent consultant anesthesiologists unconnected with the case who fill these forms after the departmental case presentation. The review forms contain comments about preanesthesia assessment, anesthesia management, and suspected reason for M&M.
We reviewed the M&M database for anesthesia for cases conducted between 1992 and 2016 (25 years) and identified all pediatric cases, age range from birth to 16 years, who had adverse events or mortality related to anesthesia. Significant adverse events were defined as either major or intermediate morbidity. We used the same definitions as used for our M&M meetings. Major morbidity was defined as a permanent disability such as cardiac arrest, neurologic or spinal cord injury. The North American Pediatric Perioperative Cardiac Arrest (POCA) Registry classifies cardiac arrest as anesthesia related if anesthesia personnel or anesthetic process played some role.5 We followed this definition. Intermediate morbidity was defined as causing serious distress or temporary but significant sequelae that resolved within 48 hours.6,7 An example of intermediate morbidity could be an unexpected admission to the intensive care unit. The reviewers also identified if anesthesia played a major role in the event in comparison to patients’ comorbidity or surgical factors.
Mortality within 48 hours of anesthesia was documented. Deaths that occurred at induction but before the start of surgery were classified as solely anesthesia-related.
We designed specific forms for the purpose of this audit and standardized the outcomes. Each form was then reviewed independently by 2 of the authors of this audit and missing details, if any, were obtained from the patients’ medical records. Each case was then discussed between the 2 reviewers to identify a major or predominant causative factor. Where this was not possible, the cases were classified as multisystem involvement. Patients undergoing cardiac surgery and anesthesia outside the OR were excluded.
Patient, anesthesiologist, and surgeon identification was not disclosed on the data collection form. Patient demographic details, type of surgery, time of events, and risk factors identified before surgery, as mentioned in the departmental review form, were noted. Adverse events were categorized and reported as cardiovascular, respiratory, central nervous system, or miscellaneous complications (which did not fit in any of the above categories).
All analyses were conducted using the Statistical Package for Social Science version 19 (SPSS Inc, Chicago, IL). Pediatric adverse events were the primary outcome. Frequency and percentage were computed for qualitative observation. The proportion of patients with adverse events and mortality, as well as 95% Poisson confidence intervals, for count data, were reported. We descriptively reported the adverse-event rate among age groups, American Society of Anesthesiologists (ASA) physical status, mode of admission (emergency versus elective), and surgical specialty.
A total of 48,828 pediatric patients underwent anesthesia between 1992 and 2016. This number did not include pediatric cardiac cases and those patients undergoing radiological procedures under general anesthesia (GA) outside the OR complex. This record was obtained from our Hospital Management Systems, where an electronic record of all cases is kept. The electronic database has been in place for 15 years. Before this, a manual record was kept.
Thirty-nine pediatric patients had significant adverse events. The demographic data, type of anesthesia, elective or emergency nature of the surgery, and surgical specialty are given in Table 1. Seventy-six adverse events were identified in 39 patients (estimated 8 patients [95% CI, 5.7–10.9] per 10,000). The systems involved and the nature and timing of events are shown in Table 2. Thirty-three percent of the patients (13 of 39) suffered respiratory events, 23% (9 of 39) had cardiovascular, and 10% (9 of 39) had neurologic events. Thirty-three percent of patients (13 of 39) had multisystem involvement and were classified as such. Hence the total number of events exceeded the number of patients.
Table 1. -
Demographic Data, Type of Procedures, and Surgical Specialty (n = 39)
| <1 mo
| 1 to <12 mo
| 1–5 y
| 6–10 y
| >10 y
|ASA physical status
|Type of surgical procedure
| Pediatric surgery
| Orthopedic surgery
| ENT surgery
Abbreviations: ASA, American Society of Anesthesiologists; ENT, ear, nose, and throat.
Table 2. -
System Involvement, Timing of Events, and Consequences to Patients (n = 39)
|Primary system involved
| Multisystem involvement
|Timing of events
| Recovery room
| Postoperative within 24 h
| Postoperative between 25 and 48 h
|Consequences to patients
| Prolonged stay in the recovery room (>2 h)
| Unanticipated admission to ICU
Abbreviation: ICU, intensive care unit.
Eleven (28.2%) of these 39 patients died within 48 hours. The estimated crude perioperative mortality rate (POMR) within 48 hours of surgery was 2.0 (95% CI, 1.1–4.0) per 10,000 anesthetics (0.02%). Seventy-three percent (8 of 11) of those who died within 48 hours had experienced an intraoperative cardiac arrest, and 34% of the 11 were neonates.
Regional technique contributed to morbidity in 4 cases. The details of these patients are provided in Supplemental Digital Content, Table 1, http://links.lww.com/AA/D185.
Thirteen children had perioperative cardiac arrest; hence, the estimated perioperative cardiac arrest rate was 2.7 per 10,000 (95% CI, 1.4–4.5) anesthetics. Fifty-four percent of these patients were infants (n = 7), 5 of whom had underlying congenital heart disease (CHD). Five of these children died within 48 hours and 1 had hypoxic brain damage. The association between CHD and cardiac arrest was an observation and was not tested statistically. In 4 of the 13 cases, the reviewers concluded that there was a major contribution from anesthesia, whereas, in the rest, there was a significant contribution from the patient’s underlying disease or surgical factors. The details of these cases are shown in Supplemental Digital Content, Table 2, http://links.lww.com/AA/D185.
In 6 additional cases, anesthesia was considered to be the predominant cause of the events that led to morbidity. The details related to the 6 additional cases are shown in Supplemental Digital Content, Table 3, http://links.lww.com/AA/D185. In total, 13 cases were identified by the reviewers, where anesthesia played a predominant role in relation to other factors present. There was only 1 death that occurred before surgery began, which was classified as solely from an anesthetic cause, that is, 0.20 per 10,000 cases (95% CI, 0.005–1.1).
In addition to the cases presented in the Supplemental Digital Content, Tables 1–3, http://links.lww.com/AA/D185, in the remaining 16 cases, anesthesia contributed partially to the morbidity in addition to surgical or patient’s coexisting disease-related factors, that is, 3.27 per 10,000 anesthetics. Anesthetic reasons in these cases included missed key information on preoperative assessment, inadequate preoperative optimization, poor communication with the surgical team or within the anesthesia team, appropriate blood products not ordered, difficulties in airway management, or decisions regarding postoperative ventilation not taken in time. The main surgical factors identified include inadequate preoperative surgical assessment, insufficient supervision by the senior surgeons, poor communication, unexpected surgical blood loss, lack of timely availability of blood products, and deficient postoperative care systems including postanesthesia care unit (PACU) and pediatric intensive care.
System errors that were identified were inadequate recovery room supervision and staffing, lack of pediatric intensive care, unavailability of a consultant in an emergency, lack of coordination, and/or inadequate preoperative assessment due to short staffing.
This review analyzed all significant pediatric perioperative adverse events and mortality that occurred in pediatric patients (newborns to 16 years) in a tertiary care institution between 1992 and 2016. All these cases were presented at our departmental M&M meetings and recorded prospectively in our departmental database. A total of 39 pediatric patients suffered a significant adverse event (8:10,000). The crude mortality rate for death within 48 hours of surgery was 2:10,000 during this period. Respiratory complications led to major M&M compared to others. A statistical comparison between those children who suffered major complications and those who underwent uneventful anesthetics was not possible, because information on ASA physical status and emergency nature of surgery was not available for the later cohort.
Other studies have shown a higher incidence of all complications in neonates and children <1 year of age.8–10 One-third of our adverse events occurred in children <1 year of age, but again we were unable to make a statistical comparison between different age groups. Although the age of patients was available for the cohort with M&M, a denominator based on differential age groups was not available from hospital records.
One of the problems in the analysis of M&M data is that there is no formal definition of major adverse events or even perioperative cardiac arrests. In spite of these inconsistencies, there is value in sharing such data and understanding the cause of these events. We used a predetermined standardized list of definitions for adverse events. Most studies on the subject have used 2 criteria for major adverse events, that is, POCA and POMR.11
Thirteen patients (33.3%) in our cohort suffered from perioperative cardiac arrest during or within 48 hours of surgery. Our overall estimated incidence of cardiac arrests was, therefore, 2.7 per 10,000 (95% CI, 1.4–4.5), and 48 hours mortality associated with cardiac arrests was 38% (n = 5).
Some of these patients were reported in our previous publication on perioperative cardiac arrests in pediatric patients between 1992 and 2006.3 The perioperative cardiac arrest rate reported in that article was 4.9 per 10,000 anesthetics. However, in that article, we only included cardiac arrests that occurred during a shorter time frame, that is, between induction until discharge from PACU. It also included 1 patient from outside OR services. Nine of those patients are included in our present data set of cardiac arrests. Between 2007 and 2016, only 3 more cardiac arrests were reported, an estimated POCA rate of 1.0 per 10,000 anesthetics. All 3 of these patients survived. The last perioperative pediatric cardiac arrest was reported in 2010. This decrease in perioperative cardiac arrests could be due to any combination of the following: introduction of mandatory pulse oximetry use, systems of pediatric anesthesia practice, safe surgery checklist use, improved surgery, and anesthesia pediatric-specific providers, and use of anesthesia drugs with a proven safety record in the pediatric population.
Fifty-four percent (n = 7) of these patients who had a cardiac arrest were <1 year of age, and 5 of the infants had underlying CHD. This association with CHD was a descriptive observation and was not tested statistically, but it agrees with other recent publications where the authors have pointed out that children with CHD are more prone to cardiac arrest associated with anesthesia when undergoing noncardiac surgical procedures.11,12
Respiratory events were the most common (30%) followed by cardiovascular events at 23%. The majority of these events occurred during the time of extubation, which included severe bronchospasm, pulmonary aspiration, and respiratory depression. Most of these events resulted in reintubation in the OR or the PACU. Additional respiratory events included failed endotracheal intubation, endobronchial intubation, incorrect ventilatory settings, fluid overload, and pneumothorax or hemothorax. Other authors have also reported respiratory problems as being more frequent in the pediatric age group compared to other systems' involvement.2,9,13 A recent systematic review recognized respiratory complications and airway problems as the most common complications.14
In 10 of the cases, anesthesia was primarily responsible (2.0 per 10,000) for morbidity. Details of all these cases are provided in Supplemental Digital Content, Table 3, http://links.lww.com/AA/D185. Determining whether anesthesia or surgery played a primary role is at times difficult; however, as stated earlier, we used the POCA registry definition.5 The rate of anesthesia mortality where the death occurred after induction of anesthesia, but before the start of surgery, was 0.2 per 10,000. This was a 16-year-old boy who had a cardiac arrest following succinylcholine at the time of induction before start of surgery. The anesthesiologist had failed to notice the high potassium values before induction in spite of the patient having acute renal shutdown. No anesthetic death solely due to anesthesia has been reported since 2006.
The anesthesia-related mortality rate in high-income countries is reported as 0.7 per 10,000 from the Netherlands15 and 0.98 per 10,000 from a tertiary care hospital in Australia.16 Data from low- and middle-income countries are scarce. Bharti et al4 from India reported pediatric anesthesia–related mortality between
April 2003 and March 2008 as 1.64 per 10,000 anesthetics where anesthesia was considered solely responsible. A Brazilian study over 5 years (2005–2010) reported no mortality due to anesthesia, and only 3 were partially related (2.81 per 10,000 anesthetics).1 Another article from a tertiary care hospital in South Africa reported their anesthesia-related mortality as 3.5 per 10,000 anesthetics with no mortality where anesthesia was solely responsible.17 The latter article also included pediatric cardiac surgery. A more recent study from South Africa reported in-hospital mortality within 24 hours of anesthesia administration in children as 16.5 per 10,000 cases (95% CI, 7.8–25.1). Fifty-seven percent were <1 year of age.18 A recent study reported on 24-hour, 48-hour, and 7-day perioperative mortality from 24 facilities in Kenya over a 36-month period between 2014 and 2016. The 48-hour mortality reported from 1 tertiary care institution in this study was 1.8%; however, the authors did not differentiate between cases where anesthesia was solely or partially responsible for deaths.19
We had 4 major incidents related to regional anesthesia but no deaths. In one incident, the child had quadriparesis following surgery where an epidural catheter was inserted for intraoperative and postoperative pain relief, and quadriparesis was noticed on the second day. The 3 other cases were related to single injection caudal with local anesthetic, and children had temporary sequelae, which included 2 bradycardic events. Bupivacaine was the local anesthetic used in all these cases. Ultrasound technique had not been used to identify the caudal space in these cases since it was not the norm at that time. Caudal anesthesia, in conjunction with GA, is commonly used in our institution, and we perform 40–50 blocks per month. Unfortunately, we do not have the total number of caudal blocks that were performed over this period; therefore, we were not able to determine the percentage of overall complications. Ropivacaine has been available since 2016 and is now preferred over bupivacaine. The rate of complications in large studies of pediatric regionals has been recorded as low. However, it has been mentioned that the rate of these complications may be underestimated.3,20
There are several limitations to this audit. Foremost is the lack of a denominator for subgroups. We were able to retrieve the total number of pediatric cases but not the proportion of neonates, infants, or other age groups. We did not have the data on the total number of regional anesthetics performed or the case breakdown according to the ASA physical status. Hence, we were not able to compare adverse events or mortality between different age groups. Although we obtained information from our M&M data, this information was limited by the amount of documentation available in the records. There is always a possibility that relevant information may have been omitted.12 Also, our database was limited to 48 hours postoperatively, whereas mortality due to anesthesia complications could have occurred beyond that period. The likelihood of underestimation in reporting, therefore, exists.
Second, data in this audit span a period of 25 years. Drugs and techniques were not standardized, and these were dependent on anesthesiologist’s preference. The medications and techniques also changed over the years, for example, replacement of halothane by sevoflurane and routine use of capnography. Third, our sample may not be representative of other large centers in the country. Our monitoring, equipment, and staff standards are better than many other institutions in the country, especially compared to the state-run institutions. However, the literature supports the publication of such data from departmental audits since they can produce important findings that cannot be obtained by other audit methodologies and can be a powerful force for change. This is the strength of this audit.
We do a yearly review of our M&M data, and it has been a positive learning experience for us. We have continuously addressed challenges in improving team communications, preoperative assessments, credentialing processes, safety standards, and have implemented system changes. This may be one of the reasons for decreasing mortality over the years. These changes have been mentioned in Supplemental Digital Content, Tables 1–3, http://links.lww.com/AA/D185.
We have no national database in Pakistan to compare with our data, and we have not found any significant pediatric perioperative data from Southeast Asia for comparison.
One of the future challenges for the countries in this region is to capture the POMR data for both pediatric and adult populations. POMR is one of the core indicators recommended by the Lancet Commission on Global Surgery for monitoring global safe surgery and anesthesia.21 Although many departments in academic institutions in the region hold some form of M&M meetings, the data are not shared or reported. This could be due to fear of litigation, practical difficulties, or lack of resources invested in this type of research. Mechanisms need to be established at national and regional levels to capture these data in the future.
Our rate of adverse events leading to intermediate and major morbidity, including perioperative cardiac arrest and mortality, was estimated as 8 patients per 10,000 anesthetics (95% CI, 5.7–10.9). Respiratory events were most common at 33%, followed by cardiac events at 23%. Thirty-three percent of the patients had multisystem involvement. Thirteen patients had a perioperative cardiac arrest (estimated 2.7:10,000), and it was observed that infants with CHD appeared to be at a higher risk. A lower rate of POCA was noticed in the last 10 years of the audit.
Twenty-eight percent of the total cohort died; hence, estimated crude POMR was 2:10,000 anesthetics. Thirty-four percent of these patients were <1 year of age.
Only 1 pure anesthetic death was reported within 48 hours (0.2:10,000 cases).
We are thankful to Amir Raza, Assistant Manager Research, Department of Anaesthesiology, Aga Khan University, Karachi, for statistical support in this study.
Name: Nasir Khoso, FCPS.
Contribution: This author helped to write the protocol, authored the first draft, and conducted revisions.
Name: Waleed B. Ghaffar, MBBS.
Contribution: This author helped to review and critique the protocol, reviewed missing information, and contributed to revision.
Name: Shemila Abassi, FCPS.
Contribution: This author helped to review and critique the protocol and drafts and contributed to revisions.
Name: Fauzia A. Khan, FRCA.
Contribution: This author helped to generate the idea, reviewed the first draft, coordinated revisions, and gave final approval of the version to be published.
This manuscript was handled by: Angela Enright, MB, FRCPC.
1. Gonzalez LP, Pignaton W, Kusano PS, Módolo NS, Braz JR, Braz LG. Anesthesia-related mortality in pediatric patients: a systematic review. Clinics (Sao Paulo). 2012;67:381–387.
2. Cohen MM, Cameron CB, Duncan PG. Pediatric anesthesia morbidity and mortality in the perioperative period. Anesth Analg. 1990;70:160–167.
3. Ahmed A, Ali M, Khan M, Khan F. Perioperative cardiac arrests in children at a university teaching hospital of a developing country over 15 years. Paediatr Anaesth. 2009;19:581–586.
4. Bharti N, Batra YK, Kaur H. Paediatric perioperative cardiac arrest and its mortality: database of a 60-month period from a tertiary care paediatric centre. Eur J Anaesthesiol. 2009;26:490–495.
5. Morray JP, Geiduschek JM, Ramamoorthy C, et al. Anesthesia-related cardiac arrest in children: initial findings of the Pediatric Perioperative Cardiac Arrest (POCA) Registry. Anesthesiology. 2000;93:6–14.
6. Haller G, Laroche T, Clergue F. Morbidity in anaesthesia: today and tomorrow. Best Pract Res Clin Anaesthesiol. 2011;25:123–132.
7. Catré D, Lopes MF, Viana JS, Cabrita AS. [Perioperative morbidity and mortality in the first year of life: a systematic review (1997-2012)]. Rev Bras Anestesiol. 2015;65:384–394.
8. Murat I, Constant I, Maud’huy H. Perioperative anaesthetic morbidity in children: a database of 24 165 anaesthetics over a 30-month period. Pediatr Anesth. 2004;14:158–166.
9. von Ungern-Sternberg BS. Respiratory complications in the pediatric postanesthesia care unit. Anesthesiol Clin. 2014;32:45–61.
10. Paterson N, Waterhouse P. Risk in pediatric anesthesia. Paediatr Anaesth. 2011;21:848–857.
11. Taylor D, Habre W. Risk associated with anesthesia for non cardiac surgery in children with congenital heart disease. Pediatr Anesth. 2019;29:426–434.
12. Flick RP, Sprung J, Harrison TE, et al. Perioperative cardiac arrests in children between 1988 and 2005 at a tertiary referral center. Anesthesiology. 2007;106:226–237.
13. Schleelein LE, Vincent AM, Jawad AF, et al. Pediatric perioperative adverse events requiring rapid response: a retrospective case–control study. Pediatr Anesth. 2016;26:734–741.
14. Bhananker SM, Ramamoorthy C, Geiduschek JM, et al. Anesthesia-related cardiac arrest in children: update from the Pediatric Perioperative Cardiac Arrest Registry. Anesth Analg. 2007;105:344–350.
15. de Bruin L, Pasma W, van der Werff DB, et al. Perioperative hospital mortality at a tertiary paediatric institution. Br J Anaesth. 2015;115:608–615.
16. van der Griend BF, Lister NA, McKenzie IM, et al. Postoperative mortality in children after 101,885 anesthetics at a tertiary pediatric hospital. Anesth Analg. 2011;112:1440–1447.
17. Cronje L. A review of paediatric anaesthetic-related mortality, serious adverse events, and critical incidents. South Afr J Anesth Analg. 2015;21:147–153.
18. Meyer HM, Thomas J, Wilson GS, de Kock M. Anesthesia-related and perioperative mortality: an audit of 8493 cases at a tertiary pediatric teaching hospital in South Africa. Pediatr Anesth. 2017;27:1021–1027.
19. Newton MW, Hurt SE, McEvoy MD, et al. Pediatric perioperative mortality in Kenya. A prospective cohort study from 24 hospitals. Anesthesiology. 2020;132:452–460.
20. Llewellyn N, Moriarty A. The national pediatric epidural audit. Paediatr Anaesth. 2007;17:520–533.
21. Meara JG, Leather AJ, Hagander L, et al. Global Surgery 2030: evidence and solutions for achieving health, welfare, and economic development. Lancet. 2015;386:569–624.