Area wise distribution and incidence
Out of 16075 entire suspected cases, 925 were positive and 15150 were negative. The overall slide positivity rate (SPR) was 5.7% in all recruited cities of Punjab however; the SPR was higher in Southern Punjab (6.1%) as compared to the Northern Punjab (5.3%). The SPR within the recruited cities of Northern Punjab was highest in Jhelum lowest in Chakwal i.e. 6.8% and 3.1% respectively. In contrast among observed cities of Southern Punjab, the SPR was maximum in Rajanpur (7.1%) minimum in Dera Ghazi Khan (5.4%). The API was 0.1 per 1000 population in all recruited cities.
Similarly, like SPR and API were also high in Southern Punjab (0.2 per 1000 population) as compared to the Northern Punjab (0.09 per 1000 population). The API within Northern Punjab was high in Jhelum and it was equal to API in Chakwal 0.4 per 1000 lowest in Gujranwala 0.04 per 1000 population. Among the analyzed cities of Southern Punjab, the API was maximum in Rajanpur and minimum in Multan i.e., 1.2 and 0.06 per 1000 population respectively. The ABER was 0.2 in Punjab and it was greater in Southern Punjab as compared to the Northern Punjab 0.26 per 1000 population > 0.2 per 1000 population. The ABER within Northern Punjab was high in Chakwal and minimum in Gujranwala i.e. 1.3% and 0.1% respectively. The ABER in Southern Punjab was highest in Rajanpur (1.7%) lowest in Multan (0.2%) (Table 1).
Seasonality variations in malaria incidence
The negative binomial regression analysis (Table 2) indicated a relationship between climate seasonality variation and malarial incidence in Punjab. Results showed the highest malaria parasitemia in Summer (June to September) followed by Autumn (October to November) and Spring (March to May). While low malaria parasitemia was observed during the peak winter dry season (December to February) (Fig. 2). The study area experiences peak rainy season in Summer (June to September). At this peak season, highest malaria parasitemia was recorded with expected log (incidence) of 0.3737 higher than that of Autumn holding other variables constant. Autumn (used as a control in the analysis) had a higher rate of incidences as compared to Spring and Winter with expected incidence (on a log scale) of -0.5671 and -2.1374, respectively.
Effect of age and gender on malaria incidence
Distribution of malaria case with respect to age and gender are given in the additional file. III. Age was a strong indicator of malaria occurrence in this study, the young people (<20 years) with 33.80% were most susceptible to malaria and risk declined gradually in older ages (Table 2). Age group<20 years is approximately 60% of total population in Punjab Pakistan and mostly involved outside activities. For example, the incidence in the age group 21-40 was -0.1271 lower (on a log scale) than that of the age group of 1-20 years holding other variables constant. Similarly, incidences in age groups 41-60 and above 60 years were -0.9392 and -2.4390 lower, respectively, then that in young people (Table 2, Fig. 3).
Males had an expected log (incidence) of 0.8466 higher than that of females holding other variables constant with 71.78% malaria cases. The P. vivax had an expected log (incidence) of 1.0308 higher than that of P. falciparum, however, mix infections are -0.8562 lower than P. falciparum (Table 2, Fig. 3).
Comparative distribution of Plasmodium species in the Northern and Southern Punjab
The species-wise distribution of malaria, according to the microscopy showed 66.7% (n=617) were P. vivax 23.7% (n=219) were P. falciparum and 9.6% (n=89) were mixed infections out of 925 recruited cases (Table 3). However, molecular analysis revealed that 53.4% (n=494) as P. vivax, 18.7% (n=173) P. falciparum and 12.7% (n=119) as mixed-species whereas 15.0% (n=139) were PCR negative cases. However, no case of P. ovale and P. malariae was found both through microscopy and PCR (Fig. 4).
In Northern Punjab74.2% (n=270) were microscopically positive for P. vivax, 19% (n=69) were P. falciparum and 6.5% (n=25) had mixed species out of 364 cases. The PCR results showed that 50.3% (n=183) were of P. vivax, 14.8% (n=54) P. falciparum, 11% (n=40) mixed species and 24% (n=87) did not amplify for any species (Table 3). Similarly, in Southern Punjab 561 cases were recruited. Out of the 347 (61.9%) were microscopically positive for P. vivax, 26.7% (n=150) P. falciparum and 11.4% (n=64) of mixed species. The PCR results showed that 55.4% (n=311) were of P. vivax, 21.2% (n=119) P. falciparum, (14.1%) (n=79), mixed species and 9.3% (n=52) PCR negative (Table 3). However, in comparison, the overall incidence was high in the Southern Punjab as compared to Northern Punjab.
The coincidental adjustment of kappa statistics specified that overall agreement in the presence or absence of Plasmodium species was good (Kappa = 0.79). However, for the detection of P. vivax the agreement between microscopy and PCR was fair (Kappa = 0.38). But for P. falciparum and mixed infection, it was moderate (Kappa = 0.53, 0.59) respectively.
Molecular Epidemiology
The average occurrence of treatment-seeking patients in all recruited cities of Punjab was 4.9%. However, the comparison of Northern and Southern Punjab indicated that the prevalence was higher in Southern Punjab (5.5%) as compared to the Northern Punjab (4.0%). Within Northern Punjab, malaria occurrence was higher in Rawalpindi i.e. 5.0% followed by Gujrat 4.9%. The malaria prevalence was lowest in Chakwal i.e. 2.3%. Among the studied cities of southern Punjab, the prevalence followed the pattern Rajanpur (6.7%) > Bahawalpur (5.3%) > Rahim Yar Khan (3.1%), Multan (2.1%) >Dera Ghazi Khan (2.9%) (Table 4).
Phylogenetic analysis
A phylogenetic tree was constructed based upon sequenced results of both P. vivax and P. falciparum taken from Northern and Southern Punjab two from each site P. vivax and one of P. falciparum (Fig. 5). The tree inferred two distinct clades. The P. vivax isolates from Northern and Southern Punjab, matched in one group, whereas all the P. falciparum isolates from the Northern and Southern Punjab in another clade. Within the main clade, P. vivax clustered into four and P. falciparum clustered into two sub-clusters. The evolutionary history was inferred using the Neighbor-Joining method. The optimal tree with the sum of branch length = 6.06796875 was shown. The percentage of replicating trees in which the associated taxa clustered together in the bootstrap test (500 replicates) was shown next to the branches. The tree was drawn to scale, with branch lengths in the same units as those of the evolutionary distances used to infer the phylogenetic tree. The evolutionary distances were computed using the p-distance method and it was in the units of the number of base differences per site. The analysis involved 21 nucleotide sequences. The codon positions included were 1st+2nd+3rd+Noncoding. All positions containing gaps and missing data were eliminated. There was a total of 85 positions in the final dataset. Evolutionary analyses were conducted in MEGA7.
One DNA sequence of P. vivax and P. falciparum showed the closest relationship with the P. falciparum partial sequence of 18S ribosomal RNA gene from Brazilian Western Amazon. Another sample of P. vivax from Northern Punjab showed the closest association with the P. vivax with a partial sequence of 18S rRNA genes from Korea. However, one sample both of P. vivax and P. falciparum from Northern and southern Punjab and one sample of P. vivax from northern Punjab showed association with P. vivax of Yunnan Province (Fig. 5).