Pre- US Canal period (1884-1904)
Despite the excellent system of hospitals and patient overall healthcare, high mortalities rates attributed to malaria were observed among French Canal employees during the unsuccessful attempt of the French companies to build a Panamá Canal between 1881 to 1889 (Figure 2A). Indeed, during the French development of the Panamá Canal a significant decline in the number of malaria cases was observed. However, efforts to control the disease during this period were highly ineffective due to the lack of information concerning malaria parasite transmission biology, particularly its transmission mode via mosquito bites. The general acceptance of the discovery proving that malaria was transmitted by mosquitoes - precisely when US took over the construction of the Panamá Canal in 1904 - had profound influence on the incidence and distribution of malaria in Panamá and the rest of the endemic countries [3-7].
1905-1956 period
Since the beginning of the Panamá Canal construction by US in 1904, and thanks to the leadership and mosquito-environmental sanitation strategy designed and “militarily” executed by colonel and physician William Gorgas; Panamá in a short time achieved a significant decrease in malaria mortality and morbidity in the two most important cities of the country at the end of the Canal: Panamá City and Colon (Figure 2A), and small areas where the Canal construction was planned. The strictly enforce integrated program involved mosquito larvae and adult control activities such as drainage, brush and grass cutting, oiling, larviciding, screening and killing of adult mosquitoes inside houses. In addition, prophylactic quinine was provided freely to all workers along the Canal construction line [7]. The death rate due to malaria in the Canal employees dropped from 11.59 per 1,000 in 1906 to 1.23 per 1,000 in 1909. [6,7]. Moreover, hospitalizations of Canal workers due to malaria gradually decreased from 82% in 1906 to 8% in 1913 [27]. Nevertheless, malaria control continued to be a challenge throughout the entire canal construction program. During that time, for logistical and economic reasons less efforts were made to control the disease in the rest of the country, particularly outside the “influence area” of this engineering work [4-6]. Consequently, between 1905 – 1931 the disease was considered prevalent throughout the country. However, besides the well-kept statistical records in Panamanian towns were US corporations of commercial interest were functioning at that time (as The United Fruit Company), data regarding the true prevalence and distribution of malaria in the entire country is incomplete [6].
In 1931, with the financial help of various organization as United Fruit Company, Gorgas Memorial Institute and the Rockefeller Foundation, a strong sanitation campaign based on pyrethroid fumigation and drainage of stagnant waters, was deployed in banana plantation towns from the interior of the country; achieving a notable decrease in the disease in these selected regions, but far from the results observed in sanitized areas of the Canal Zone [6]. It was estimated that by 1947 malaria morbidity in the Panamanian population was around 8441 patients, and that before the beginning of the Global Malaria Elimination Programme in 1956, 2849 malaria cases were counted [28]. However, these figures should be viewed with caution because reporting of malaria cases and deaths was not mandatory in the country until 1957.
Between 1931 to 1949 the predominant species causing malaria in Panamá was P. falciparum [29-31]. In a study conducted by Clark and Komp (1938) it was found by microscopic assessment that P. falciparum infections reached 73.2%, P. vivax 13.9%, P. malariae 1.5%, mixed infections 9.7%. As for mixed infections, it was observed that those by P. vivax predominated with P. falciparum [30]. It was noted, however, that with the introduction of DDT (Dichloro-Diphenyl Tricloethane) in the country for mosquito control in 1947, co-infections and re-infections with P. falciparum gradually disappeared, with a concomitant increase in cases due to P. vivax [28]. This change was most likely due to the distinct characteristics of the biology of P. vivax and the behavior of its insect vector, which makes P. vivax malaria difficult to control. Among these specific features, the presence of P. vivax hypnozoites which can reactivate weeks or months after the primary infection, is one of the major reasons of the predominance of this species after indoor residual spraying (IRS) was installed in the country.
During this period (1905-1956) important discoveries regarding methods for malaria control and treatment were developed and evaluated on the isthmus in seminal longitudinal studies carried out in endemic communities from Panamá [27, 29-32, 33-42].
1957 – 1999 period
Historical trends of reported malaria cases and the most relevant events that have influenced malaria dynamics in Panamá between 1957 - 2019 are summarized in Figure 2B.
In 1956, following WHO and PAHO guidelines, the Global Malaria Eradication Programme (MEP) was launched in Panamá along with the rest of Central America and Mexico. That next year the official registration and the mandatory notification of malaria cases in the country also began. The Programme at that time operated under a vertical structure and was nationally disaggregated in operation areas [43,44]. The first report in 1957 accounted for 7,361 malaria cases mostly by P. vivax, 186 deaths and an API of 8.1 per thousand inhabitants (Figure 2B, Additional files 1 and 2).
In the years following the MEP there was a marked decline in malaria cases, although evaluations carried out in 1960 proved that transmission remained active in all the country provinces [45]. In 1958 Dieldrin was the insecticide solely used for IRS throughout the country, with annual periodicity cycles. In 1962 Dieldrin was replaced by DDT with semiannual cycles per year, obtaining great success in the campaign [45,46]. This change in insecticide was for economical and logistical reasons, not because of an evidenced Anopheles resistance to Dieldrin in the country. Furthermore, Dieldrin was considered a highly toxic insecticide and there was reluctance to its application because, according to the residents, it killed their domestic animals. There were also reports of An. albimanus resistance to Dieldrin in El Salvador [46,47].
Within the next few years following the eradication campaign, the number of cases continued to drop, but not to the figures expected by the MEP [45]. This decline trend in morbidity was observed until 1966, when malaria reached alarming values, reaching 3639 cases and an API of 3.0 (Figure 2B, Additional files 1 and 2).
With the creation of the MoH in 1969, the progress of the Elimination Programme became one of the national health authorities’ priorities. However, the statistical data for that year were discouraging; malaria incidence raised to 5906 cases with 24 deaths [47]. Moreover, the API reached 4.4 per thousand inhabitants, the highest since 1960 and the greatest magnitude observed to the present (Figure 2B, Additional files 1 and 2). Around 90% of the cases registered that year where from the provinces of Panamá, Colón and Darién (Figure 1). Field studies carried out in 1970 confirmed for the first-time clinical resistance of P. falciparum to chloroquine (CQ) in some locations from Panamá Province [48]. Thus, CQ was replaced by sulfadimethoxine associated with pyrimethamine and used for radical treatment in those areas. In general, the period between 1957 - 1970 was marked by political disturbance and a consequently deterioration on the MEP. Thus, it was the period with greatest morbidity and mortality observed in the history of malaria in Panamá (Figure 2B).
Between 1971 and 1972 resistance of A. albimanus to DDT was detected in various regions of the country and was replaced by the carbamate Propoxur for IRS in areas where DDT resistance was verified [52,53]. By 1972 the last report of P. malarie infection occurred [54], without additional cases reported to date. The following year (1973) international funding for the MEP concluded and thus the program started to be exclusively financed by national funds. Consequently, in that same year there were difficulties in materials supplies, deficiencies in supervision and in implementing control activities in remote areas [51]; a situation that promoted malaria epidemics in several regions of the country totaling 1,588 cases (Figure 2B). The most affected areas were in the eastern provinces: Darién and San Blas (now known as Comarca Guna Yala), near the Colombian border (Figure 1).
The period from 1975 to 1985 represented the lowest malaria incidence in Panamá since the MEP was created in 1957, with a focalized transmission of malaria (Figure 2B). However, in the late 1980s (1985-1989) Panamá went through a serious political crisis accompanied by a similar economic recession that culminated in the military invasion by US in 1989 [10]. This crisis had a profound effect in the NMCP activities. There were deficiencies to cover all expenses, mainly for the purchase of insecticides and to cover operating costs that would allow adequate IRS coverage [48]. The effects were rapidly felt on the declining case trend observed during the previous years (Figure 2B). Between 1986 – 1988 an annual average of 1,000 cases was registered, mainly due to P. vivax epidemics in indigenous remote communities from the Province of Darién and the Eastern region from Panamá province, accounting up to 95.0% of the total cases registered in the country during that period. Given vector resistance and the attributed detrimental effects to health, in 1988 the use of DDT for vector control was banned and replaced by carbamate (Sumithion 40% WP or Sumithion 50% EC depending on the household physical characteristics) and organophosphate (Propoxur) insecticides. In 1993, pyrethroids (Cyfluthrin, Solfac Deltamethrin, and K-othrine) were evaluated, but at that time were not implemented by the NMP as alternatives for IRS [49]. In 1996, deltamethrin was reevaluated and that same year it began to be used, replacing Sumithion. This change represented important savings for the NMP since deltamethrin was applied twice per year whilst Sumithion cycle was three times a year. However, deltamethrin was discontinued in 2002 and replace by fenitrotion, after the resistance of An. Albimanus to this insecticide was detected.
In the period between 1990 - 2000 the epidemiological pattern varied from year to year, with an average number of cases 725 ± 161 and an API in the range from 0.2 to 0.4 per 1,000 inhabitants (Figure 2B, Additional files 1 and 2).
2000 – 2019 period
By the year 2000, malaria morbidity rate reached 36.5 per thousand inhabitants in the country and P. vivax was responsible for 96% of infections. Two important issues regarding malaria control took place in Panamá at the beginning of the millennium. First, the country joined the Rollback Malaria (RBM) strategy proposed by WHO, that focused more in control than elimination of the disease [50]. Second, following international guidelines the MoH completed the process of decentralization of the malaria program. Two years later, in 2002, the malaria reemergence was declared in Panamá with 2244 cases and an incidence of 75.7 per 100,000 (Figure 2B). This number represented a 2.4-fold higher relative risk compared with the incidence observed in 2001, the previous year. Furthermore, it was the highest incidence in the last 27 years, only comparable with the one observed in 1974 (73 per 100,000). Not only malaria risk increased in 2002, but also the disease significantly spread throughout the country. Making things worse, in 2003 autochthonous P. falciparum transmission resumed in Kuna Yala and Eastern Panamá, a situation not observed since 1970. It was also observed that circulating P. falciparum parasites in Panamá presented mutations that conferred resistance to chloroquine and partial resistance to antifolates, precisely the first and second line antimalaria drugs used to treat P. falciparum cases by the NMP at that time [55,56]. These relevant resistance findings were later confirmed using molecular barcode assays developed for P. falciparum [57].
The situation continued to deteriorate reaching a peak of 5094 malaria cases and six officially recorded deaths in 2004; figures only comparable to what occurred in the country in the late 1960s (Figure 2B).
To tackle this public health crisis a Vector Control Task Group was created by the MoH. For this purpose, a crisis budget was allocated to this Group to be exclusively used for malaria control activities, without intromission by any other entities from the administrative structure. All staff and resources of the program were placed under the coordination of the Program supervisor and an intensive operational plan was established to guarantee the following activities in remote areas: an active surveillance, rapid outbreak containment and a high IRS coverage. In this way, only by means of administrative modifications without changes in the attention guidelines, four months after the Vector Control Task Group creation a significant drop in the incidence rate was observed (Figure 2B). At the end of 2005, the API was 1.4, a value that represented a 29% decrease compare with the previous year (API = 1.7). This decreasing trend continued with a 70% reduction (API = 0.5) in 2006 and a 76% reduction (API = 0.4) in 2007, reaching 354 cases and an API = 0.1 in 2011; the lowest incidence since 1985 (Additional file 2). Additionally, the autochthonous transmission of P. falciparum was eliminated in Guna Yala and the mortality rate significantly decreased.
However, in 2012 when the country was in a sustained control phase, a rebound in malaria transmission was observed, doubling the number of cases (354 and 860) from the previous year. Unfortunately, from 2013 to 2019 the number of malaria annual cases have remained above 500, reaching a peak of more than 1,400 cases in 2019. More importantly, since 2015 P falciparum transmission has reestablished in eastern regions of the country.
Regarding regional strategies and commitments to eliminate malaria, Panamá did not meet the goal of 75% reduction of malaria morbidity at the end of 2014 set at the 58th World Health Assembly using as baseline estimated cases for 2000 [54]; and most likely, Panamá will not achieve the goal to eliminate local malaria cases by 2020 established by NMEP in 2016 [12].
Malaria distribution by age, sex, and geographic location; 2000-2019
In general, from 2000 to 2019, Panamá has accumulated a total of 28,921 infected people and 25 deaths from malaria. In this period morbidity has been more frequently observed in men than in women (58.8% vs 41.2%; p < 0.001). We also found significant differences when comparing malaria case among age groups (p < 0.001) (Figure 3). Half of the infected population was 19 years old or younger, and 41% corresponded to children under fifteen years (9297/22,712), especially infants up to five years. In men, 75% of the cases occurred among those who were below 35 years of age, while in women the age was 32 years or less (Figure 3). Additionally, malaria has progressively increased its prevalence within indigenous settlements. In 2005, 41.8% of the total malaria cases were from indigenous communities, while in 2016 this proportion reached 84.6% and in 2019 more than 90%. In fact, more than 70% of the cases accumulated in the country since 2005 come from indigenous communities located in the East of Panamá (Figure 4).
Imported Malaria 2000-2019
The Panamanian - Colombian border represents an important and continuous threat to accomplish the elimination goal set by the NMEP. Indeed, more than 14,000 migrants crossed into Panamá illegally from Colombia between January and June of 2019 [58]. Around 55% were from the Caribbean, primarily from Haiti and Cuba, 25% from Africa, 19% from Asia and the rest from South America. Between 2000 and 2019, 361 malaria imported cases have arrived from different regions of the world. Most cases were from the American region 81.6% (294/360), particularly Colombia (48.6%; 175/360), Costa Rica (18.8%; 68/360), Nicaragua (3.8%; 14/360) and Venezuela (3.6%; 13/360). Countries from the African Region (13.3%; 48/360) and from Southern Eurasia (5.2%; 18/360) also contributed with a significant percentage of imported malaria in Panamanian territory (Figure 5). Of the imported cases during this period, 28.5% (103/361) were P. falciparum and 71.5% (258/361) P. vivax. The burden of imported malaria due to P. falciparum has been led mainly by Colombia (55.3%) and ten countries from the African continent (38.8%) (Figure 5, Additional files 3 and 4).
Association between malaria incidence and ENSO events
Annual malaria cases recorded in Panamá between 1957 – 2019, with years classified according to different ENSO phases is shown in Figure 6A. Figure 6B shows potential breakpoints in malaria transmission that occurred in 1968, 1970, 2002 and 2005. In Table 1 AIC values are indicated for models that split the time series in 5, 4 and 3 segments and for a null model without segments; the best model included the following 4-time segments 1957-1970,1971-2002,2003-2005,2006-2018. Parameter estimates for this best model are shown in Table 2. On average for the period 1957-1970 there were 3364 malaria cases per year during normal ENSO phase years, a number that increased by 37 and 38% during the Cold and Hot ENSO phases respectively. For the 1971-2002 period, the number of malaria cases significantly decreased by 83% when compared with 1957-1970 (P<0.05), while for 2002-2005 it increased by 17% when compared with 1957-1970, although not significantly. This corresponds to the time when malaria transmission increased following the decentralization of the NMP. From 2006 to 2018, the number of malaria cases significantly decreased by 79% when compared with 1957-1970 (P<0.05), to a level similar to what was observed between 1971 and 2002.
Table 1 Akaike information criterion for negative binomial models explaining the number of malaria cases through different time segments.
|
No. Segments
|
Segments
|
AIC
|
5
|
1957-1968,1969-1970,1971-2002,2003-2005,2006-2018
|
988.33
|
4
|
1957-1970,1971-2002,2003-2005,2006-2018
|
986.46
|
4
|
1957-1968,1969-2002,2003-2005,2006-2018
|
1021.8
|
4
|
1957-1968,1969-1970,1971-2005,2006-2018
|
1028.3
|
4
|
1957-1968,1969-1970,1971-2002,2003-2018
|
1017.5
|
3
|
1957-2002,2003-2005,2006-2018
|
1055
|
3
|
1957-1970,1971-2005,2006-2018
|
1027
|
3
|
1957-1970,1971-2002,2003--2018
|
1015.7
|
1
|
1957-2018
|
1063.1
|
Table 2 Parameter estimates for the best negative binomial model explaining the number of malaria cases as function of the transmission time segment and ENSO phase.
|
Parameter
|
Case number change
|
Estimate
|
Std. Error
|
z value
|
Pr(>|z|)
|
ENSO-Normal 1957-1970
|
3364.383
|
8.121
|
0.15
|
54.152
|
<2e-16*
|
1971-2001
|
0.17
|
-1.7713
|
0.1645
|
-10.767
|
<2e-16*
|
2002-2005
|
1.174
|
0.1602
|
0.3199
|
0.501
|
0.6165
|
2006-2018
|
0.206
|
-1.58
|
0.19
|
-8.316
|
<2e-16*
|
ENSO-Cold
|
1.374
|
0.318
|
0.1683
|
1.889
|
0.0589
|
ENSO-Hot
|
1.383
|
0.3246
|
0.148
|
2.193
|
0.0283*
|
Overdispersion
|
3.985
|
0.686
|
-
|
-
|
-
|
*Statistically significant (P<0.05)
|
|
|
|
|