Study Design

We performed a descriptive cross-sectional study. Residents and cardiac surgeons were contacted by email from the LACES database and invited to answer a voluntary survey. Information was anonymous. The survey was conducted in English. Given the small number of surgeons in some countries, demographic information, such as age, could only be answered in the survey as ranges (e.g., 40-50 years), but not with an exact value, thus ensuring the data anonymity. Hence, exact median or mean values cannot be calculated. The questionnaire was constructed considering the following domains: training, research and academic participation, professional satisfaction, financial retribution, and workload. Questions to address each of these domains were constructed. A pilot study was performed among random surgeons and residents to ensure the questionnaire’s comprehensiveness and obtain feedback regarding additional or irrelevant questions. Reliability was assessed based on internal consistency evaluation. The survey was divided into three parts: 1) general questions for all participants, 2) questions only for trainees (residents/fellows), and 3) questions only for staf (graduated) surgeons. The survey can be seen in the Supplementary Appendix. Answers were collected from March to April 2021. Participants from the following countries answered the survey (Figure 1): Argentina, Aruba, Bolivia, Brazil, Chile, Colombia, Costa Rica, Cuba, Dominican Republic, Ecuador, Guatemala, Honduras, Mexico, Panama, Paraguay, Peru, Uruguay, and Venezuela.

Fig. 1 - Countries of origin of eligible survey respondents.

Download

Fig. 1 - Countries of origin of eligible survey respondents.

Statistical Analysis

The normality of data was assessed using the Shapiro–Wilk test. Normally distributed data were presented as mean ± standard deviation. Non-normally distributed data were presented as median with interquartile ranges. Non-parametric variables were assessed and compared using Student’s t-test or chi-squared test, as appropriate. All data were tabulated, and analyses were performed using Microsoft® Excel® (Microsoft, Redmond, Washington, United States of America) and Prism™ (GraphPad LLC, San Diego, California, United States of America). Statistical significance was accepted at P<0.05. To avoid violating confdentiality and ensure the anonymity of the respondents given the small number of surgeons in some participating countries, some results (e.g., age) are limited to ranges rather than expressed as specific numeric values.

Respondents’ Demographics

Thirty-seven responses were excluded based on country. After excluding ineligible responses, analyses were performed on the remaining (N=289) responses. Respondents originated from 18 different countries (Figure 1) with a median age between 40-50 years. Respondents’ demographics are summarized in Table 1.

Table 1 - Demographics of survey respondents.

Work and Leadership

Most respondents reported working 40-80 hours (N=175, 60.6%) or 80-120 hours (N=71, 24.6%) per week. Few reported working < 40 hours per week (N=26, 9.0%). Weekly working hours were significantly higher for residents and fellows compared to surgeons (P<0.001).

The majority (N=205, 70.9%) of respondents reported choosing cardiac surgery as their specialty of training again if given the option, whereas 23.5% (N=68) were unsure, and 5.5% (N=16) would prefer a different specialty. Among surgeons, 72.4% (N=160) would choose cardiac surgery again, whereas 5.9% (N=13) would not. Among residents and fellows, 66.2% (N=45) reported choosing cardiac surgery again, whereas 4.4% (N=3) would not.

Forty-one (14.2%) respondents stated that leadership positions are widely available and accessible in their institution, city, and/or country. In contrast, 62.6% (N=181) respondents stated that it was difficult to obtain a leadership position in some capacity. The remaining respondents did not consider leadership positions, either due to lack of interest (N=28, 9.7%), lack of time (N=6, 2.1%), or having to take up too many responsibilities with little or no pay (N=27, 9.3%).

Surgeon-Specific Responses

A total of 221 surgeons (76.5%) responded, of whom 9.5% (N=21) were female and 90.0% (N=199) were male. Respondents stemmed from 16 different countries with a median age between 40-50 years. Among the surgeons that responded, 91.4% (N=202) completed surgeon-specific questions.

Most surgeons worked in public centers (N=111, 55.0% vs. N=91, 45.0% in private centers) and completed fellowship training (N=147, 65.1% vs. N=74, 34.9%). The most common fellowships pursued were congenital heart surgery (N=40, 18.9%), heart transplantation (N=18, 8.5%), and aortic surgery (N=17, 8.0%). Transcatheter interventions fellowships and minimally invasive cardiac surgery fellowships were completed by 18 (8.5%) and 15 (7.1%) respondents, respectively. Annual cardiac surgery case volumes were evenly distributed (Figure 2). Two-thirds of surgeons trained residents (N=139, 68.8%), and nearly three-quarters of them (N=144, 71.3%) performed research to some extent.

Fig. 2 - Annual cardiac surgical case volume by staf (graduated) surgeon respondents (N=202).

Download

Fig. 2 - Annual cardiac surgical case volume by staf (graduated) surgeon respondents (N=202).

Surgeons’ training ranged from less than four years (N=51, 25.2%) to more than eight years (N=15, 7.4%), with a median of 5-6 years (Table 2). Also, 18.8% (N=38) reported not having undergone general surgery training, while most conducted either 1-2 years (N=65, 32.2%) or 2-4 years of general surgery training (N=68, 33.7%), with a median of 1-2 years. The median cardiac surgery training surgeons underwent was 3-4 years, ranging from 1-2 years (N=8, 4.0%) to more than four years (N=88, 44.2%). Critical care training varied widely from having received no dedicated training (N=56, 27.8%) to more than two years of training (N=62, 30.8%), with a median of 3-6 months. Most of surgeons (N=129, 63.9%) indicated that establishing a regional cardiac surgery board exam would be beneficial, while 5.9% (N=12) were completely against it. Three-quarters of surgeons (N=149, 73.8%) stated that it was difficult to find a cardiac surgery job after completing training.

Table 2 - Training durations reported by cardiac surgeons, residents, and fellows.

Surgeons’ monthly salaries widely varied, with a median between 2,500-5,000 USD (or United States Dollar). Most surgeons reported being unsatisfied (N=92, 45.3%) or moderately satisfied (N=86, 42.4%) with their salary. Most surgeons (N=165, 82.1%) reported having received a salary during their residency training. Nearly one in fve surgeons (N=41, 20.4%) stated they had to pay for residency training, of whom 73.2% (N=30) had to take loans to aford to pay for residency.

Resident/Fellow-Specific Responses

Sixty-eight residents and fellows (23.5%) responded (23.7%, N=15 female vs. 76.3%, N=53 male). Respondents had 14 different countries of origin with a median age between 30-40 years. All residents and fellows completed all or most resident/fellow-specific questions.

The anticipated median duration of training for residents and fellows is expected to be 5-6 years. Residents and fellows completed a median of less than one year of general surgery training and less than three months of critical care medicine training and are expected to complete a median of more than four years of cardiac surgery training (Table 2). Most residents and fellows received a salary during training (N=58, 85.3%). One in fve (N=13, 19.1%) reported having to pay for residency training.

The comfort levels of residents and fellows in different clinical and surgical tasks varied considerably (Figure 3). Most (N=41, 60.3%) reported independently performing less than one case per week, on average, whereas contributions to essential parts of operations vary (N=21, 30.9% less than one per week; N=13, 19.1% one per week; N=22, 32.4% 2-5 per week; N=12, 17.6% more than fve per week). A third (N=23, 33.8%) of residents and fellows anticipated having performed less than ten procedures independently by the end of their training compared with 30.9% (N=21) expecting to have performed more than 50 independent procedures. Half of the respondents (N=32, 47.1%) reported that their program had all resident spots filled, whereas 11.8% (N=8) had one open spot, 17.6% (N=12) had two open spots, and 23.5% (N=16) had more than two resident spots unfilled. A median of 2-4 night shifts were performed per week, with 8.8% (N=6) reporting 5-7 night shifts per week. Satisfaction with training programs varied, with 22.1% (N=15) being satisfied, 48.5% (N=33) moderately satisfied, and 29.4% (N=20) unsatisfied.

Fig. 3 - Residents’ and fellows’ level of comfort in performing procedures. CT=computed tomography; ECG=electrocardiogram; ICU=intensive care unit; MRI=magnetic resonance imaging.

Download

Fig. 3 - Residents’ and fellows’ level of comfort in performing procedures. CT=computed tomography; ECG=electrocardiogram; ICU=intensive care unit; MRI=magnetic resonance imaging.

Most respondents (N=61, 89.7%) stated they wish to pursue or are currently pursuing fellowship training. The most reported fellowships being pursued or expected to pursue included minimally invasive cardiac surgery (N=18, 26.5%), aortic surgery (N=11, 16.2%), congenital heart surgery (N=12, 17.6%), and transcatheter interventions (N=10, 14.7%). Most of residents and fellows (N=52, 76.5%) were in favor of the development of a regional cardiac surgery board exam, whereas 2.9% (N=2) were opposed. Research was commonly but variably performed during training (N=48, 70.6%).

Limitations

The main limitation of this study is that the survey results might not necessarily represent the complete picture and reality of LATAM since 289 respondents are only a small fraction of the total cardiac surgery trainees and staf surgeons in LATAM. Moreover, other important topics/questions might still be missing and were not included in the survey to increase the survey completion rate since otherwise, the survey might be too long, and respondents would tend to abort the questionnaire. Finally, to avoid violating confdentiality and ensure the anonymity of the respondents, some results are limited to ranges rather than expressed as specific numeric values, thus reducing the specificity of some results.