Annals of Clinical Microbiology and Antimicrobials Open Access Prevalence and Antimicrobial Resistance Pattern of Bacterial Meningitis in Egypt

Infectious diseases are the leading cause of morbidity and mortality in the developing world. In Egypt bacterial diseases constitute a great burden, with several particular bacteria sustaining the leading role of multiple serious infections. This article addresses profound bacterial agents causing a wide array of infections including but not limited to pneumonia and meningitis. The epidemiology of such infectious diseases and the prevalence of Streptococcus pneumoniae, Neisseria meningitidis and Haemophilus influenzae are reviewed in the context of bacterial meningitis. We address prevalent serotypes in Egypt, antimicrobial resistance patterns and efficacy of vaccines to emphasize the importance of periodic surveillance for appropriate preventive and treatment strategies.


Introduction
Emerging infectious diseases will always pose a world threat because of the continuous battle with pathogens that undergo antigenic changes to escape our immune system, and resist antimicrobial treatment. Communicable diseases are particularly challenging in developing countries because of poor-socioeconomic conditions that facilitate the spread of the pathogens and the abuse of antimicrobial therapy that results in emerging antimicrobial resistant strains, that are resistant to conventional antimicrobial treatments and possibly vaccines. This article addresses communicable diseases that are either transmitted through airborne mechanisms or contact including droplets and discharge from nose and throat. The epidemiology of three bacterial meningitis agents will be reviewed with emphasis on prevalence, antibiotic resistance and serotypes, to provide the platform for effective treatment and vaccine strategies in the region. Strepto-coccus pneumoniae is a leading causative agent of diverse infections. In Egypt, it was recently described as the leading cause of bacterial meningitis [1,2] skewing the epidemiology from Neisseria meningitidis, which was previously reported as the major etiological agent [3][4][5][6]. Acute respiratory infections caused by S. pneumoniae are frequently reported, yet more recent data are needed to form a comprehensive and updated understanding of serotype distribution and antimicrobial resistance pattern in the region [1,7]. A few studies addressed pulmonary infection from S. pneumoniae [8,9], with the majority of the studies addressing pneumococcal meningitis. Reports on the serotypes of isolates obtained during the studies of pneumococcal meningitis represent a fraction of all the serotypes causing different forms of pneumococcal diseases [7]. Additionally, the prevalent serotype distribution had shown variation along different studies conducted at different time intervals [7,10,11]. In order to assess effec-tively the epidemiology of the diseases for effective preventive and treatment strategies we need to periodically reassess the serotype prevalence. Currently the major serotypes reported in Egypt are {6B,1,19A,23F,6A} which are inadequately represented in the current 7-and 11valent vaccine [2,7] urging the production of an effective regional vaccine. Additionally, penicillin resistance was reported by several studies at different time intervals [1,2,8,9,12], with an increase in the pattern of resistance over time; in 1993 71% S. pneumoniae were susceptible to penicillin [8], in 2000 63% of isolates were susceptible to penicillin [9], in 2004 51% of isolates were susceptible to penicillin [2], therefore periodic monitoring of the patterns of antimicrobial resistance is necessary to guide effective treatment [4]. In a study conducted from 1998-2004, 4% of S. pneumoniae (of 560) isolates conferred multidrug resistance and 50% of these were characterized as serotypes 23F, 6B, and 6A, which are the prevalent serotypes in Egypt [2]. A recent study conducted from 1998-2003 to identify the antimicrobial susceptibility and serotype distribution focused on pneumococcal isolates obtained from CSF [7]. However, isolates causing different forms of pneumococcal diseases other than meningitis were not addressed.
Neisseria meningitidis was for long reported as the leading cause of bacterial meningitis in Egypt, since the early studies conducted in 1965-1989 [3,5,13]. Recently it was described as the second or third leading cause after S. pneumoniae [1,2,4]. Earlier studies have reported serogroup A to be the predominant virulent serogroup [3,13,14]. However, few studies indicated that serogroups B [1,2,4] and C are replacing serogroup A [10,15], and this change in the epidemiology of meningococcal meningitis can be attributed to the use of the group A polysaccharide vaccine [16][17][18]. Unfortunately, case mortality in groups B and C were reported to be higher than A [3,10]. Sulpharesistant N. meningitidis was detected since 1965 [5,6]. Due to the lack of standard criteria to evaluate the antibiotic susceptibility of N. meningitidis, isolates are not evaluated to determine the resistance in most studies. To the best of our knowledge, Afif et al., was the only study that addressed the resistance profile findings in Egypt where 86% of isolates were found resistant to co-trimoxazole and more than 40% were resistant to penicillin [2]. Most recently a molecular study investigated culture negative CSF specimens from meningitis patients to rigorously identify prevalent etiological agent, they reported an increase in detection rate of N. meningitidis from 16% to 23% making it the second leading cause of bacterial meningitis [2]. It was rationalized that the presence of the high number of culture negative disease in several studies conducted in Egypt may be due to the uncontrolled use of antibiotics prior to hospital admission [2,4,6,10,19].
Haemophilus influenzae was less implicated in meningitis in earlier studies conducted in Egypt [3,13], yet recently it was reported (serotype b) to be one of the leading causes of bacterial meningitis in children [1,2,4]. Resistance to ampicillin, chloramphenicol, ceftriaxone and co-trimoxazole were detected [1,2]. It was reported in several studies that serotype b is responsible for invasive diseases caused by H. influenzae [4,5,10]. However, in a study conducted in 1993 on Egyptian children below 5 years H. influenzae non-b was found to contribute to 43% of the pneumonia isolates, questioning the large scale use of Hib vaccine. Sparse epidemiology studies are available addressing the role of H. influenzae as a pathogen in Egypt [20].
The epidemiology of bacterial meningitis will be addressed in the following section by reviewing the prevalence of each of these bacteria over 39 years, and addressing antimicrobial resistance, serotypes dominance and vaccine efficacy.

Streptococcus pneumoniae
Streptococcus pneumoniae is known to give rise to several severe infections. In Egypt it was recently described as the leading cause of bacterial meningitis [1,2,4] reflecting a change in the epidemiology of the disease where N. meningitidis was for a long time the main etiological agent causing bacterial meningitis [3,5,6,13]. Acute respiratory infections caused by S. pneumoniae require extensive reporting [1, 7,8]. More recent data need to be generated pertaining to serotype distribution and antimicrobial resistance patterns, to ensure effective treatment measures. The following paragraphs address reports on meningitis caused by S. pneumoniae in Egypt.

Pneumococcal meningitis is currently the leading cause of meningitis in Egypt
Meningitis caused by S. pneumoniae is often referred to as pneumococcal meningitis. S. pneumoniae has one of the highest mortality rates amongst meningitis cases especially in patients less than one year of age [3,10,13].
Several studies were conducted between 1965 and 2004 on the epidemiology of pneumococcal meningitis, and revealed a constant rise in the number of S. pneumoniae meningitis in Egypt. All studies conducted before the mid 1990s demonstrated S. pneumoniae either at second or third place as the meningitis-causing agent in Egypt [3,5,6,13]. A longer comprehensive study ES 1966-1989 on 7,809 patients admitted to the Abbassia Fever hospital (AFH), reported that 7.3% of patients that suffered from meningitis were due to pneumococcal infection and the peak in the number of cases were during Jan-April. The mean age of the patients was 11.7 with 41% mortality, this is five times greater than the mortality caused by N. meningitidis which was the leading cause of bacterial meningitis during this period. Expectedly, 68% of the mortality cases reported were less than one year of age [3].
A one year study ES 1977-1978 conducted on 1627 CSF specimens obtained from two fever hospitals in Cairo revealed 350 specimens positive for bacterial infection, where pneumococci were the prevalent bacteria detected compromising 8.7% of the total specimens studied (1627) and 40.6% of bacterial positive specimens (350) with most cases reported from Jan-May. It is worth noting that the etiological agent for a large number of the specimens was not identified. The mortality rate was 44% and again mortality was mainly reported in patients less than 1 year. Serotyping by Quellung reaction identified type 1 to be the most capsular type [10].
In 2000 a project performed by the Ministry of health and population (MOHP) was carried out in 12 hospitals and identified S. pneumoniae as the leading cause of bacterial meningitis. A total of 2455 persons were suspected with acute meningitis where an overall of 223 had acute bacterial meningitis (ABM). S. pneumoniae was identified in 32% of patients with ABM [1]. Additionally, severe cases, based on clinical criteria of ABM were highly reported with S. pneumoniae, and were the etiological agent of 46.5% of the winter cases [21]. An articulate molecular study that included 14 hospitals in Egypt to determine the epidemiology of bacterial meningitis in 11,070 patients suspected with the disease during ES 1998-2004 identified S. pneumoniae to be the leading cause, responsible for 42% of the 843 culture-positive bacterial meningitis cases, and 6% of 1,784 culture-negative CSF specimens tested by PCR. This study by Afifi et. al., used molecular tools to address the etiological agent of the disease when conventional methods failed. In other studies when culture yields negative results the disease was described as purulent meningitis [13], without further investigation of the causative agent [2]. The high percentage of culture negative samples was reported in many studies in Egypt. This was explained by the high frequency of patients receiving on Percentage of pneumococcal meningitis cases based on studies conducted over 39 years Figure 1 Percentage of pneumococcal meningitis cases based on studies conducted over 39 years. A representation of nine different studies conducted on patients diagnosed with meningitis due to different etiological agents from 1965-2004. In the study ES 1965-1968 on 644 cases of ABM 3% (21 patients) of the cases reported were due to pneumococcal infection [5]. ES [1966][1967][1968] conducted on 187 meningitis patients S. pneumoniae was responsible for 29% of 123 culture-positive cases [6]. ES 1966-1989 conducted on 7,809 meningitis or encephalitis patients, reported that 7.3% of total cases were due to pneumococcal infection [3]. A retrospective review ES 1971-1975 of 1,333 patients with ABM, reported 7.9% of meningitis cases diagnosed with pneumococcal meningitis [13]. ES 1977-1978 revealed pneumococci was the prevalent bacteria detected [10]. Surveillance of 2047 patients with meningitis in ES 1998-2000 was conducted to determine the etiology of the disease in children less than 6 years. 30% of 228 patients of the cases confirmed by culture were positive for S. pneumoniae. However, in children older than 12 months and less than 6 years, S. pneumoniae was the leading causative agent [4]. ES 2000 revealed that S. pneumoniae compromised 32% of ABM [1]. ES 2002-2003 on 310 children clinically diagnosed with meningitis; 202 patients were diagnosed with ABM; 13.9% of total meningitis cases were infected with S. pneumoniae and 21.3% of ABM [21]. ES 1998-2004 S. pneumoniae was responsible for 42% culture-positive bacterial meningitis cases [2]. The asterisks represent epidemiological studies reporting an average during the entire period of the study. the counter antimicrobial drugs prior to professional evaluation. [2,4,6,10,19]. Figure 1 summarizes percentage of pneumococcal meningitis cases based on studies conducted over 39 years.

Antimicrobial resistance and treatment
Extensive studies are required to continually update antimicrobial susceptibility patterns. In developing countries the unregulated use of antibiotics is not uncommon, stressing the importance of surveillance of antibiotic resistant pathogen to guide empirical treatment. Several studies reported penicillin resistant S. pneumoniae [1,2,8,9,12], showing an increase in resistance pattern over time. Table 1 shows the decrease in the percentage of penicillin susceptible S. pneumoniae from 1993-2004 [2,8,9]. Such information on patterns of antimicrobial resistance is an effective mean to guide rational treatment [4].
One of the initial studies ES 1991-1993 that determined the resistance pattern of S. pneumoniae was conducted in Abbassia and Embaba fever hospitals. The study involved 1635 children; 961 patients isolates of S. pneumoniae were recovered from nasopharynx and blood. Table 1 illustrates the resistance pattern of the blood isolates tested in this study, it is important to note that the nasopharyngeal isolates had similar susceptibility patterns to those of the blood. The study revealed a low percentage of antimicrobial resistance [8]. An increase in penicillin resistance was recorded in a later study conducted by MOHP where 0.8% resistance was detected, however all isolates were susceptible to vancomycin [1]. In 1998-2000 the antimicrobial resistance pattern reported (refer to table 1) recommended ceftriaxone as the drug of choice for treatment of children with bacterial meningitis [4].
A retrospective multicenter study ES 1999-2000 conducted in 5 hospitals in Egypt revealed an increase in penicillin resistance, and little resistance to Ceftriaxone (84% susceptible) and Ciprofloxacin (82% susceptible) [9]. A recent sentinel meningitis surveillance program ES 1998-2003 showed a marked increase in penicillin resistance (50%-100/205) among CSF isolates in Egypt [7]. Afifi et al., reported high rates of multidrug resistance in S. pneumoniae; 4% of the 206 isolates tested, urging the need to control the dispensing of antibiotic which are generally The percentages of non-susceptible S. pneumoniae isolates to eight conventional antibiotics to S. pneumoniae isolated from CSF and/or blood in eight studies conducted from 1991-2004 are summarized. The percentage of penicillin susceptible S. pneumoniae dropped between 1993-2004; from 71% S. pneumoniae penicillin susceptibility in 1993 [8], 63% of isolates were susceptible to penicillin in 2000 [9], to 51% penicillin susceptibility in 2004 (2). The study conducted in 1998-2000 recommended ceftriaxone as the drug of choice for treatment of children with bacterial meningitis [4].
(Intermediate resistance is abbreviated as I, and Resistance as R) available as an over-the-counter medication [2]. Additionally, a surveillance report of the ARMed (Antibiotic Resistance Surveillance & Control in the Mediterranean Region) project which started in 2003 and continued for 2 years in the southeastern Mediterranean, reported 30% penicillin resistance and 25% erythromycin resistance among the S. pneumoniae Egypt isolates [12,22].

Changes in prevalence serotype
In order to assess the efficacy of vaccine and to provide the foundations for the development of effective vaccines for the region it is critical to determine the serotype distribution in Egypt. However, a caveat in such assessment is that isolates tested in the epidemiological studies were from pneumococcal meningitis patients and therefore the identified serotype might not necessary represent all the serotypes causing different forms of pneumococcal diseases as certain serotypes may have a propensity to invade one clinical site. Additionally, variation in the prevalent serotype amongst different studies conducted at different time intervals were shown [7,10,11]. ES 1998-2003 study conducted to identify the serotype distribution addressed only pneumococcal isolates obtained from CSF and therefore may not necessarily reflect isolates causing different forms of pneumococcal diseases other than meningitis [7]. A study conducted in Egypt in the late 1970s used Quellung reaction, and identified type 1 as the most frequently observed capsular type [10]. The study by Wasfy et al., during ES 1998-2003 reported the major serotypes {6B, 1, 19A, 23F, 6A} from CSF specimens of patients with meningitis [7]. ES 1998-2004 study by Afifi et al., confirmed these finding, the study was conducted during the same time period, and found the same predominant serotypes in Egyptian patients [2]. Additionally 50% of multidrug resistant isolates were characteristic of serotypes 23F, 6B and 6A [2]. These serotypes are inadequately represented in the current 7-(4, 6B, 9V, 14, 18C, 19F, 23F) and 11valent vaccine (contain four additional serotypes 1, 5, 3 and 7F) and therefore non of the current vaccines are used in Egypt [2,7]. The predominant serotypes reveal enhanced penicillin resistance when compared with other serotypes. Interestingly, Guirguis et al., reported in 1990 that serotypes 1, 6A, 9L, 12A, 19A and 29 are the most frequent serotypes identified in the study [11]. These studies impose the importance and urgency to periodically collect and analyze more data on the current predominant serotypes present in Egypt [2]. This will aid in effective treatment strategies, effective vaccine administration and effective means to synthesize new effective vaccines for the region. During the period from 1966-1989 the country experienced 3 outbreaks with a periodicity ranging from 6-8 years [23].

Neisseria meningitidis
Gradual decrease in the prevalence of Meningococcal meningitis Meningitis caused by Neisseria meningitidis is referred to as meningococcal meningitis. In early studies from 1965-1968 meningococcus was responsible for more than half (56%) of the cases of bacterial meningitis patients (figure 2). The most affected age group was between 5-15 and the highest mortality cases were reported in ages less than one year. During this 4 years study there was a continuous rise in death rate, which was explained by the development of meningococcal strains resistant to sulpha [5]. Further studies confirmed the prevalence of N. meningitidis as the leading cause of bacterial meningitis [3,6,13,23].
N. meningitidis possessed a pattern in the occurrence of outbreaks "cyclic nature"; a gradual increase in the number of meningococcal cases every 6-8 years [3,23]. Serogroups A, B, and C were all reported however group A constituted the majority of cases [3,13,14,23]. Most cases occurred during winter months (Jan to Apr) in patients between the ages of 5 and 14. Serogroup A had the least mortality rate compared to serogroups B and C. Mortality was highest in patients less than 1 year and more than 24 years [3]. During the last 2 years of ES  an observed change in the clinical presentation of patients with meningococcal meningitis was reported where the onset of the disease was more acute and the course was more severe. Petechial rash was more common in the last years with a higher percentage of mortality occurring in patients developing the skin rash [3,23].
ES 1977-1978 studied 1627 suspected meningitis cases and performed CSF analysis; 5.5% of the total suspected cases and 25.4% of the 350 culture-positive meningitis cases were caused by N. meningitidis. This was reported as the second most common bacterial meningitis with a mortality rate of 21% [10]. This was the only study during this period that identified serotype C as the predominant serogroup followed by B then A. Most cases occurred from November to April affecting mainly infants and children. Meningococcal meningitis cases had the lowest overall mortality when compared with other bacterial meningitis infections, yet its mortality rate was higher compared to previous studies [10] possibly due to the prevalence of serotype C [3,10].
meningitidis was reported to be responsible for 13% of the 228 bacterial meningitis cases with 23% case fatality. This study reported serogroup B being more common than serogroup A, however, it is worth noting that this was based on 8 isolates tested [4]. A Laboratory-based surveillance of patients with bacterial meningitis was conducted in 14 hospitals from 1998-2004 to determine the etiology and the antimicrobial susceptibility of meningitis pathogens. N. meningitidis was responsible for 17% of the 843 cultured positive cases placing it third place after S. pneumoniae and H. influenzae. When PCR was performed on culture negative isolates the N. meningitidis became the second leading cause constituting 23% of the cases [2]. These studies show a gradual decrease in the prevalence of Meningococcal meningitis and Figure 2 illustrates the percentage of meningiococcal meningitis cases from 1965 to 2004.

Antimicrobial resistance and treatment
Resistance of meningococcus to sulphonamides in Egypt was detected in early studies in the 1960s [5,6] and   Table 2 summarizes N. meningitidis resistance to penicillin, ampicillin, chloramphenicol, ceftriaxone, tetracycline, and trimethoprim/sulfamethoxazole from 1998-2004, note that N. meningitidis were isolated from CSF and/or blood.

Shift in the serogroups distribution following the implementation of the bivalent A/C vaccine
Early studies indicated that the major serogroup in Egypt was serogroup A, which accounts for 95% of the cases [3,13,14,17]. The observed cyclic nature of the disease reported in the ES 1966-1989 study permitted the design of the first successful serogroup A meningococcal polysaccharide vaccine trial in Egypt [3,18]. The results of the first trial conducted in ES [1971][1972] indicated that the vaccine conferred significant protection where it was able to produce high degree of immunity for the following three years when compared to the control cases [18]. In 1977 the second clinical assessment for serogroup A meningococcal vaccine was carried out in Alexandria on school children to assess the effectiveness of the vaccine, it was found to be effective for one year following immunization. There was an observed decrease in protection after the first year, which was inconsistent with what was observed during the first trial [16]. A study conducted during the same period 1971-1972 by Sippel et al., revealed a high number of carriers of group B and C, anticipating a possible change in the epidemiological pattern of the serogroups in Egypt [15]. A study by Guirguis et al., ES 1977 reported an unusual pattern in serogroups where C was the predominant followed by B then A [10].
In 1992 the MOHP initiated a school based vaccination program with the bivalent A/C capsular polysaccharide vaccine. The vaccine was taken at age of 6 years followed by a second dose 3 years later. A retrospective review by Nakhla et al., to understand the change in epidemiology of meningococcal serogroups after the introduction of the bivalent vaccine indicated a significant decrease in meningococcal cases following the implementation of the school-based vaccination program [17]. It is worth noting that the expected outbreak in 1996 has actually never occurred and it was reported that there is a decline in the incidence of meningococcal meningitis cases [23,24].
A more recent study ES 1998-2000 reported a higher number of serogroup B than A [4]. Another study by Afifi et al., reported out of 135 isolates of N. meningitidis 51% were group B, 35% group A, 4% group W-135 and 2% belonged to group D and Y [2]. The project conducted by MOHP in 2000 in addressing the communicable diseases in Egypt reported that meningococci were compromised of 54.5% of serogroup B, 31.8% group A, 4.5% group W135, 2.3% group Y [1]. This shift in the serogroups was observed after the implementation of the bivalent A/C vaccine by MOHP implying the significant impact that the vaccine had on the epidemiology of the disease [17].

Haemophilus influenzae
Sparse epidemiology studies are available addressing the pathogenicity of H. influenzae. It was less implicated in meningitis in earlier studies conducted in Egypt [3,13], yet recently it was reported in limited studies to be one of the leading causes of bacterial meningitis in Egyptian children [1,2,4]. It is difficult to estimate the incidence of H. influenzae meningitis in Egypt due to absence of population based analysis and limited surveillance studies [4]. One study ES [1998][1999][2000][2001][2002][2003][2004] , defined H. influenzae serotype b as the main cause of pneumonia in children below 5 year [2].
H. influenzae the leading cause of meningitis in very young children A prospective study ES 1966-1989 on 7,809 patients admitted to the Abbassia Fever hospital, reported H. influenzae to be the cause of meningitis in 322 of the patients (4.1%) who had a mean age of 2.5 years; it is worth noting that most  [21]. In a laboratory-based surveillance ES 1998-2004 carried out to identify the etiological agent of bacterial meningitis, H. influenzae was found to be the second leading cause following S. pneumoniae responsible for 20% of the 843 cases of culture positive patients. Among children less than 5 year H. influenzae was the most common bacteria isolated. The mean age was 7 months with a case fatality ratio of 25% [2]. Figure  3 illustrates the prevalence and trends of H. influenzae in bacterial meningitis cases.

Antibiotic resistance and treatment
The study by Ostroff et. al (table 3). The yield of isolates from blood as indicated in table 3 was very low (only 6) to be able to draw definitive conclusions regarding the resistance pattern [8].
The Surveillance conducted by MOHP reported increase in the resistance pattern from the above study. It also reported multi-drug resistance (TET, SXT, and CHL) in 29.6% of the isolates, in addition to positive β-lactamase activity [1].
Youssef et. al., demonstrated that less than 50% of H. influenzae isolates were susceptible to chloramphenicol and ampicillin while all isolates were susceptible to third generation cephalosporins; ceftriaxone. It was reported that higher mortality rate were found in patients with H. influenzae resistant isolates to AMP and CHL [4]. In the laboratory-based surveillance ES 1998-2004 conducted in a network of 14 hospitals 45% ampicillin resistant isolates were reported, 37% of which demonstrated beta-lactamase production (Table 3) [2].

Serotype distribution and vaccination
Since earlier studies H. influenzae serotype b was predominantly reported among the few reported cases. Guirguis et al., reported that all H. influenzae isolates were serotype b; out of 47 H. influenzae isolates 30 were tested and all of them were type b [4]. The finding that the majority of H. influenzae isolates were type b [20] and a common cause of acute lower respiratory infection allowed the introduction of Hib vaccine by the MOHP in the EPI (Expanded Program on Immunization) [1,4].

Concluding Remarks
The dominating causative agent for bacterial meningitis has changed over the forty years period studied. Studies have demonstrated that Pneumococcal meningitis is currently the leading cause of meningitis in Egypt. Studies on the prevalent bacterium in different age group shows that H. influenzae is the leading cause of meningitis in very young children. The serotype prevalence of some bacterial species were altered over the years changing the mortality rate of the disease. Several bacteria have developed antimicrobial resistance to conventional antibiotic regimes and emerging multidrug resistance strains. The referenced studies have provided the platform for assessing the prevalence and antimicrobial resistance pattern of bacterial meningitis in Egypt. However, this also iterates the need for periodic and continual surveillance to guide the empirical treatment of such diseases.

Competing interests
The authors declare that they have no competing interests.