Introduction

SARS-CoV-2 was discovered as the etiological agent of the Wuhan pandemic ^1-3. Faced with the spread of the virus, Peru opted for mandatory social distancing (border closures, mandatory social isolation and ban on the use of transportation nationwide), but the evolution was not as expected. By the 130th day of lockdown, the number of infected persons had risen to 375,961 and the case fatality rate (CFR) had increased to 4.75% ^4,5.

In this context, a heterogeneous progression of the disease was observed within the different regions of Peru. For example, by 2020, in epidemiological week 22 (EW 22), Huancavelica had 245 cases and a CFR of 0.41 % ⁵, while other regions reported higher rates. Although there are publications reporting a reproduction rate (R0) of 2.97 for Peru and 2.88 for Lima ⁶, it is not clear how the number of infected persons has evolved during the pandemic in high-altitude areas.

Initially, in the absence of effective treatments or safe and effective immunization, non-pharmacological measures (NPMs) were chosen in Peru to "mitigate" the transmission of SARS-CoV-2 ^7-10. In a given context, traditional public health measures such as isolation, lockdown, social distancing and reduction of economic activities are considered as NPMs. However, in Peru, both the efficacy and efficiency of NPMs were affected by the non-compliance of the population ¹¹, the low response capacity of health centers or the early recovery of economic activities.

The present research reports the clinical and epidemiological characteristics of COVID-19 patients in an area of Huancavelica, in the Peruvian high-altitude region, and details an approximation of the impact of NPMs on the transmission of SARS-CoV-2.

Materials and methods

Results

Out of 1,454 records of COVID-19 patients diagnosed at Hospital de Lircay II-1, 50 were removed. Out of these, 48 were deleted because they had two positive tests on different dates and two were eliminated because they died at home with a diagnosis of COVID-19 post-mortem and without previous clinical care; therefore, 1,404 records were left (n = 1404). On the other hand, only 129 records included symptoms and date of symptom onset.

Table 1 presents the epidemiological and clinical characteristics of the study subjects, whose profile consisted of a median age of 33.0 (IQR: 24.0-45.0); age range of 30 to 59 years (49 %, n = 688); female predominance (59 %, n = 828); no comorbidity (73.2 %, n = 1,028); reactive IgM and IgG diagnostic tests (53.3 %, n = 749); headache, general malaise and cough as main symptoms; clinical classification with predominance of asymptomatic patients (n = 1,091), followed by mild (n = 276) and moderate-severe (n = 37) cases.

Table 1 and Figure 1 show an average CFR of 0.9 % (n = 13), a higher percentage of moderate-severe clinical cases (24.3 %, n = 9) and a highly affected elderly population.

Figure 1 COVID-19 CFR according to the age of diagnosed cases, Hospital de Lircay II-1, Angaraes, Huancavelica, Peru (March 31, 2020- January 31, 2021) 

The transmissibility of the virus throughout the study was determined by the Rt (Figure 2). It revealed an estimated Rt of 0.88 as of July 12 (95 % CI: 0.26-2.10). In contrast, the beginning of the progressive economic recovery, around July 1, caused an increase in the transmissibility of the virus, with the Rt increasing up to 2.98 (95 % CI: 1.54-5.14) on July 17 and up to 4.72 (95 % CI: 2.93-7.12) on August 1. On the contrary, focused lockdown resulted in a decrease in the Rt to 0.79 (95 % CI: 0.41-1.36), two weeks after the implementation of said measure. Likewise, during this period, the transmissibility rate remained below 1. However, once again, the lifting of the lockdown and some restrictions caused an increase in the fluctuating Rt.

Figure 2 Rt for the COVID-19 pandemic according to diagnosed cases, Hospital de Lircay II-1, Angaraes, Huancavelica, Peru (March 31, 2020-January 22, 2021). The shaded area represents the 95 % confidence interval for the median. The numbers on the dotted lines represent the following: 1: June 30, 2020; 2: July 31, 2020; 3: September 18, 2020; 4: December 1, 2020; and 5: December 23, 2020. 

Discussion

The COVID-19 population of Lircay, at 3,278 m a.s.l., had epidemiological characteristics similar to those reported in other studies; however, they also showed important clinical differences. Early publications from China indicated that the infection was common in adult and elderly patients, with ages ranging from 30 to 79 ¹ years and median ages of 49 ¹⁶ and 47 ¹⁷ years. In the present study, 82.4 % of the COVID-19 patients aged between 18 and 59 years with a median age of 33 years. A study ¹⁸ conducted in Huaraz, Peru, at 3,052 m a.s.l. found that COVID-19 patients had a median age of 41 years, while in another report ¹⁹ from Huancayo, Peru, at 3,250 m a.s.l., the mean age was 49 years. The difference between these findings was probably due to the demographic distribution of the population in the high Andean region.

In the present study, COVID-19 in symptomatic patients was characterized by general malaise, headache and cough. Early reports in China indicated that cough was a frequent symptom in patients with SARS-CoV-2 infection ^16,17; however, its frequency was slightly exceeded by general malaise (53.5 %) and headache (51.2 %) ^16,17. Other publications indicate that, in high Andean areas of the country, 49.5 % ¹⁸ to 70.26 % ¹⁹ of the symptomatic patients had cough. The discrepancy with other national publications could be due to the fact that, in this study, 77 % of the cases were classified as asymptomatic, which could have caused a bias by missing data in MINSA’s statistical computer system.

The average CFR of this study was 0.9 %, which is lower than that reported for the first COVID-19 cases in China. Thus, since the outbreak in Wuham, CFRs were low and in the range of 1.4 % ¹⁷ to 2.3 % ¹⁶. In Peru, Díaz-Lazo et al. ¹⁹ reported a rate of 14.21 in the high altitude region of Huancayo, Peru; meanwhile, Flores et al. ²⁰ found a CFR of 32 per 100,000 inhabitants in the department of Huancavelica between March and September 2020.

On the other hand, Ortiz-Prado et al. ²¹ claimed that, in Ecuador, in regions below 2,500 m a.s.l., mortality rates were 24 % higher than at higher altitudes. In contrast, the CFR in hospitalized patients was 18.27 % in Santiago, Chile ²²; 12.7 % in Fortaleza, Brazil ²³; and only 5 % in Buenos Aires, Argentina ²⁴.

The CFR of the present study was considerably increased in the most severe cases (CFR of 24.3 % for moderate- severe cases). Previous reports in hospitalized Peruvian patients from sea level regions show slightly higher CFRs, from 29.4 % ²⁵ to 32.8 % ²⁶. Apparently, the lethality of the virus is lower in this high Andean region, but it should be considered that not all the inhabitants were treated at Hospital de Lircay II-1.

Although a lower COVID-19 CFR at a higher altitude could be plausible, such results—and especially those of this study—could be conditioned by factors such as the characteristics of the population (lower age and low prevalence of comorbidities), demographic factors (low population density) and lesser or greater social mobility or other measures (greater or lesser use of masks, use of non-conventional medicines). In addition, part of the study population was treated in other conventional health systems (EsSalud and private health centers) and possibly in non-conventional systems.

An asymptomatic patient was one whose diagnosis was confirmed by a COVID-19 test, being RT-PCR the most commonly employed ^27-29. A report of absence of symptoms during a follow-up period (5 days before to 14 days after testing) was prepared ²⁷, which was enough to differentiate asymptomatic from presymptomatic cases. In this research, it was found that 77.7 % of COVID-19 cases in Lircay reported no symptoms at diagnosis. According to some studies, the percentage of asymptomatic infections accounted for 17 % ²⁷ to 32% ³⁰; however, research conducted in Cuba reported up to 80 % ³¹ while in Peruvian sea-level regions 25.7 % ³². The high percentage of asymptomatic patients reported in the present study is probably due to the fact that, at the time of the research, an "asymptomatic case" was confirmed, in some cases, by a molecular test and, in others, by a positive COVID-19 antigen test or IgM or IgM/IgG reactive serology test ^33,34.

The present research has used the Rt estimation as well as previous research carried out in China ^15,10 to evaluate the evolution of the pandemic in Lircay. It was found that the early national economic recovery and the lifting of the national lockdown, around the beginning of July 2020, and the start of the year-end holidays shopping (December 2020) increased SARS-CoV-2 transmissibility. The increased number of cases after the economic recovery, promoted by the Peruvian Government, did not consider the heterogeneous spread of the virus. By the beginning of July 2020, at the national level, the "wave" of infections seemed to be decreasing, which would have motivated the lifting of the most severe restrictions.

The results of this study show that, during the pre-economic recovery phase, diagnosed cases were low and did not reach a peak in diagnosis rate. Thus, a low number of infections would lead to a high number of susceptible people who, in addition to the lifting of the national lockdown, would have led to an increase in the transmissibility of the virus.

Focused lockdown was established as a pandemic control measure, which was evidenced by a decreased Rt. The progressive lifting of mandatory social isolation, associated with social distancing measures and reduced capacity of public places, kept the Rt below 1 during November.

Paradoxically, after the increase in transmissibility at the beginning of December, the Rt decreased at the beginning of the holidays around December 24.

A similar phenomenon was observed in the study by Pan et al. ¹⁰: during the first period of the outbreak in Wuhan (January 1 to 10, 2020), due to the Chinese Lunar New Year mass travel, the Rt dropped even though there was no restriction of any kind. In addition, the total number of cases reported in each province in China was associated with the total number of travelers leaving Wuhan ^35-37. Thus, the decreased transmissibility in Lircay at the end of December could possibly have been caused by the migration of people from the district to other regions to celebrate the year-end holidays, with the resulting theoretical export of COVID-19 cases to the destination sites.

One of the main limitations of the study was the use of the SISCOVID-19 database as a source of information. Although it is a great online resource that helped to integrate data at the national level, the epidemiological records were incomplete and most of the information was obtained from such database. Likewise, those files only included the results of serology or antigen tests, since the results of molecular tests (RTPCR) were recorded on another online platform and the number of such tests was very small.

Another limitation was the lack of certainty about the characteristics of other non-pharmacological interventions, such as the use of barriers like masks and face masks, or non-conventional interventions specific to the regions in the Peruvian inland. Extrapolation to the total population of Lircay is impossible due to the certainty that the supply of care was also covered by EsSalud and, probably, by private health centers.

Finally, although the number of comorbidities in the study was low, there is no certainty about the presence or absence of communicable and non-communicable conditions specific to the region, or pathologies related to high altitude, such as chronic mountain sickness, or other cardiovascular conditions that are more frequent at higher altitude levels.