Morbidity,mortality and antimicrobial resistance of pneumococcal infections in the Jamaican paediatric and adult populations

Abstract

BACKGROUND:

Pneumococcal infections are a leading global cause of morbidity and mortality, complicated by the increasing antimicrobial resistance of pneumococcal isolates.

OBJECTIVE:

To evaluate morbidity and mortality associated with both invasive pneumococcal disease (IPD) and non-IPD in Jamaica in both the paediatric and adult population. Pneumococcal isolates ( $n=$ 94) were collected over a 2-year period (2008–2009).

METHODS:

Risk factors for poor clinical outcomes: death, complicated disease and length of hospitalization (LOH) were evaluated and antimicrobial resistance patterns were determined by Kirby-Bauer disc diffusion.

RESULTS:

The case fatality rate was 6.8%. Independent mortality risk factors included complicated disease [OR 30.9 (3.4–276.6)] and diabetes mellitus [OR 8.3 (1.4–48.8)]. Independent risk factors for the development of complicated disease included sickle cell disease [OR 36.5 (4.2–320.3)] and sepsis [OR 3.5 (1.2–10.4)]. The LOH was increased most in patients with invasive disease (4.6-fold) and resistance to ceftriaxone (4.3-fold). Penicillin (16.0%) and erythromycin (14.9%) resistance was most prevalent, while ceftriaxone (4.3%) resistance was least prevalent.

CONCLUSIONS:

The high burden of IPD in at-risk groups in our population and the associated increase in morbidity and mortality underlie the need for improved preventive and therapeutic management strategies in these patients.

Keywords

Pneumococcus streptococcus pneumoniae morbidity mortality jamaica

1. Introduction

Streptococcus pneumoniae is the most common cause of acute otitis media, sinusitis, pneumonia and one of the most important causes of bacterial meningitis worldwide [10, 15]. The overall global morbidity and mortality of invasive disease caused by this pathogen, particularly pneumonia and meningitis, have remained high despite the availability of effective antimicrobial therapy [8]. The case fatality rates for pneumococcal pneumonia in particular, which ranges from 7 to 36%, has remained largely unchanged for the last 50 years [5]. Management of these infections are further complicated by the increasing resistance of the isolates to antibiotics often used in management of these patients [3].

The pathogenesis of pneumococcal disease is the result of a delicately balanced series of complex and as yet incompletely understood interactions between host and bacteria [12]. Pneumococcal colonization, which is most prevalent in the paediatric population, serves as a reservoir for disease through translocation into sterile sites like the bloodstream and meninges or inter-person droplet transmission [12]. An increased risk of invasive pneumococcal disease (IPD) development is associated with defects in the non-specific or specific immune defense mechanisms against colonization, aspiration or invasion of S. pneumoniae [12, 15]. Such underlying conditions that affect immune function include HIV infection, congenital or acquired asplenia and concurrent respiratory infections with influenza virus, Haemophilus influenzae and Mycoplasma pneumoniae. In studies of IPD worldwide, predisposing conditions can be found in approximately 22% to 43% of patients [9].

There is an urgent need for accurate data on the morbidity and mortality associated with both IPD and non-IPD in Jamaica in both the paediatric and adult population. The results of this study will serve to guide the implementation of appropriate preventive and therapeutic management strategies for various groups at risk for developing pneumococcal disease in Jamaica.

2. Patients/methods

2.1 Subjects and bacterial strains

The study was carried out between 2008 and 2009, recruiting patients presenting at the University Hospital of the West Indies (UHWI), the major tertiary referral center in Kingston and St. Andrew in the eastern region of the island. The pneumococcal isolates ( $n=$ 94) recovered from clinical specimens collected from patients were collected consecutively over the study period and only one isolate per patient per infectious episode was included in the study. Ethical approval of the study was obtained from the UHWI institutional oversight committee.

2.2 Demographic and clinical data

A standardized data abstraction form was used to collect demographic and clinical information from hospital clinical records. The data collected pertained to age, gender, type of clinical specimen and date of collection, clinical presentation, morbidity, mortality and the presence of any co-morbid illnesses that increase risk for pneumococcal disease. Invasive disease was defined as cases in which pneumococci was isolated from a normally sterile body site such as the blood stream, central nervous system or synovial fluid [17]. The primary clinical outcome measure was death from any cause and secondary outcomes included development of complications and the length of hospitalization (LOH). The development of complications was defined as the need for intensive cardiovascular or respiratory resuscitation measures. The LOH was calculated in days from initial clinical assessment to discharge from the hospital and the LOH for decedents was not included in association analyses including this variable.

2.3 Microbiology laboratory procedures

Standard microbiological procedures were followed and bacterial isolates were identified as S. pneumoniae by microscopy, colony morphology, positive optochin sensitivity and bile solubility tests [18]. The sensitivity of pneumococcal isolates to penicillin (P), ampicillin (AMP), amoxicillin-clavulanate (AMC), ceftriazone (CRO), trimethoprim-sulfamethoxazole (SXT) and erythromycin (E) was determined by the Kirby-Bauer disc diffusion method. A control pneumococcal strain (ATCC49619) was utilized during microbiological testing of samples. The pneumococcal strains were stored in tryptic soy broth containing 15% glycerol at $-$ 70 ${}^{\circ}$ C until required [2].

Table 1
Clinical characteristics of patients presenting with pneumococcal disease

Clinical characteristic	All patients ( $n=$ 94)	Paediatric ( $<$ 18 yo) ( $n=$ 46)	Adult ( $\geqslant$ 18 yo) ( $n=$ 46)
Age (mean $\pm$ SD [range])*	28.0 $\pm$ 28.3 [NB to 86]	2.8 $\pm$ 3.2 [NB to 13]	52.5 $\pm$ 18.9 [18 to 86]
Age (median (IQR] years)*	18.0 [2.0–51.5]	2.0 [0.7–4.0]	51.5 [38.3–67.8]
Sex (male)	53 (56.3)	30 (65.2)	23 (50.0)
Invasive disease	65 (69.1)	22 (47.8)	41 (89.1)
Sepsis	49 (52.1)	20 (43.4)	28 (60.9)
Pneumonia (LRTI)	36 (38.2)	7 (15.2)	27 (58.7)
Co-morbid illness	32 (34.0)	6 (13.0)	25 (54.3)
HIV	7 (7.4)	1 (2.2)	6 (13.0)
SCD	7 (7.4)	2 (4.3)	4 (8.7)
DM	7 (7.4)	0 (0.0)	7 (15.2)
Solid tumor	4 (4.3)	1 (2.2)	3 (6.5)
Haem. malignancy	2 (2.1)	0 (0.0)	2 (4.3)
COPD	2 (2.1)	0 (0.0)	2 (4.3)
Malnutrition	1 (1.1)	1 (2.2)	0 (0.0)
Viral pneumonia	1 (1.1)	1 (2.2)	0 (0.0)
Extensive burns	1 (1.1)	0 (0.0)	1 (2.2)
Conjunctivitis/periorbital cellulitis	15 (16.0)	15 (32.6)	0 (0.0)
URTI	8 (8.5)	7 (15.2)	1 (2.2)
SSI	6 (6.4)	0 (0.0)	6 (13.0)
CNS infections	5 (5.3)	2 (4.3)	3 (6.5)
Otitis	2 (2.1)	2 (4.3)	0 (0.0)

COPD – chronic obstructive pulmonary disease, DM – diabetes mellitus, HIV – human immunodeficiency virus, NB – newborn, SCD – sickle cell disease, URTI – upper respiratory tract infections. *Age data was not available for 2 individuals.

Figure 1.

Clinical presentation of pneumococcal infections among various age groups. CNS – meningitis and intracerebral abscess cases, Ear – otitis cases, Eye – conjunctivitis and/or periorbital cellulitis cases, LRTI – pneumonia cases, SSI – skin and skin structure infections, URTI – upper respiratory tract infections. The 2 cases in which age data were not available were necessarily excluded from this analysis.

2.4 Statistical methods

Continuous data variables were expressed as mean $\pm$ standard deviation or median with the corresponding interquartile range. Univariate association between dichotomous variables was determined using Pearson chi-squared or Fisher’s exact test as appropriate and the strength of association was expressed as odds ratios with 95% confidence intervals (OR 95% CI). The significance of difference in means was determined by the independent t-test with a two-tailed test of significance. For multivariate analyses, a forward stepwise linear regression model and forward stepwise logistic regression model were used for continuous and dichotomous response variables respectively. The analysis was performed using the xlstat software, Version 16.6 (Addinsoft, NY, USA) and results were considered statistically significant if p values were less than 0.05 ( $p<$ 0.05).

3. Results

3.1 Clinical presentations

Sepsis with or without a known primary focus of infection (52.1%), pneumonia (38.2%) and conjunctivitis (16.0%) accounted for the majority of clinical presentations (Table 1). The majority of these sepsis cases occurred in children younger than 5 years (38.3%) and adults greater than 60 years (27.7%) (Fig. 1). Almost half of sepsis cases occurred without an identified primary infective focus (22/49, 44.9%). In descending order, bacteremia was most prevalent in cases of CNS infections (60.0%), pneumonia (58.3%) and skin and skin structure infections (SSI) (33.3%) but absent in other patients. Most otitis (100%) and conjunctivitis cases (86.7%) occurred in children below 5 years, all in children below 18 years. In contrast, CNS (60.0%), pneumococcal pneumonia (76.4%) and SSI (100%) cases occurred most frequently in adults over 18 years.

Risk factors for pneumococcal disease were present in 34.0% (32/94) cases. These included human immunodeficiency virus infection or acquired immune deficiency syndrome (HIV/AIDS) (7/94, 7.4%), sickle cell disease (SCD) (7/94, 7.4%), diabetes mellitus (DM) (7/94, 7.4%), solid tumors (4/94, 4.3%), hematological malignancy (2/94, 2.1%), chronic obstructive pulmonary disease (2/94, 2.1%), malnutrition (1/94, 1.1%), viral pneumonia (1/94, 1.1%) and extensive burns (1/94, 1.1%). Most of the cases of HIV and DM occurred in the adult population while SCD was present in both the paediatric and adult populations.

Table 2
Clinical summary of the fatal cases of pneumococcal disease in the UHWI cohort

Age (years)	Clinical diagnosis	Gender	LOH (days)	Cause of death	Co-morbid illness
0.67	Sepsis	Male	0.7	Hepatic/Splenic sequestration, sepsis	SCD
48	Sepsis	Female	31	Thoracoabdominal aortic aneurysm, sepsis	HIV/AIDS
64	SSI	Male	16	Multiorgan system failure, sepsis	None
65	LRTI	Male	17	ARDS/MI	DM
66	CNS infection	Female	1	Meningitis, uncontrolled DM, sepsis	DM
73	CNS infection	Female	0.5	Meningitis, pericarditis, bronchopneumonia	None

AIDS: Acquired immunodeficiency disease, ARDS: Acute respiratory distress syndrome, HIV: Human immunodeficiency virus, LOH: Length of hospitalization from intake interview to demise, MI: Myocardial infarction.

Table 3

Risk factors associated with clinical outcome measures

Outcome/risk factors	Estimated risk	$p$ -value
Death	OR (95% CI)
Complicated disease	30.9 (3.4–276.6)	$<$ 0.0001
Age $>$ 60 years	8.8 (1.7–44.9)	0.006
DM	8.3 (1.4–48.8)	0.013
Complicated disease	OR (95% CI)
SCD	36.5 (4.2–320.3)	$<$ 0.0001
Invasive disease	5.6 (1.4–22.6)	0.016
Co-morbid illness	3.5 (1.3–9.5)	0.011
Sepsis	3.5 (1.2–10.4)	0.021
Conjunctivitis	0.2 (0.1–0.4)	$<$ 0.0001
Mean LOH	Mean LOH $\pm$ SD (days) in affected vs unaffected/fold increase
Invasive disease	12.8 $\pm$ 11.6 vs 2.8 $\pm$ 4.7/4.6-fold	$<$ 0.0001
CRO resistance	35.7 $\pm$ 25.5 vs 8.3 $\pm$ 8.9/4.3-fold	$<$ 0.0001
Sepsis	14.4 $\pm$ 12.5 vs 4.6 $\pm$ 6.3/3.1-fold	$<$ 0.0001
CNS infections	23.5 $\pm$ 16.7 vs 8.5 $\pm$ 10.1/2.8-fold	0.006
AMP resistance	20.3 $\pm$ 24.3 vs 8.3 $\pm$ 4.7/2.4-fold	0.022
AMC resistance	20.8 $\pm$ 23.9 vs 8.9 $\pm$ 9.2/2.3-fold	0.036
Complicated disease	15.2 $\pm$ 10.1 vs 7.9 $\pm$ 10.6/1.9-fold	0.021
Conjunctivitis	1.3 $\pm$ 3.0 vs 11.0 $\pm$ 11.2/0.11-fold*	$<$ 0.0001

* Note that this translates to a roughly 8-fold decrease in LOH. AMC: Amoxicillin-clavulanate, AMP: Ampicillin, CNS: Central nervous system, CRO: Ceftriaxone, DM: Diabetes mellitus, LOH: Length of hospitalization, OR (95% CI): Odds ratio (95% confidence interval), SCD: Sickle cell disease.

3.2 Antimicrobial resistance and pneumococcal morbidity and mortality

In decreasing order, isolates were most resistant to penicillin (15/94, 16.0%), erythromycin (14.9%), trimethroprim/sulfamethoxazole (9.6%), ampicillin(5.3%), amoxicillin/clavulanate (4.3%) and ceftriaxone (4.3%).

A total of 6 patients died during the course of their illness giving a case fatality rate of 6.8% (6/94). The majority of these deaths (3/6, 50.0%) occurred during the first day; the remaining 3 patients dying on days 16, 17 and 31 of their illness. All but one of these patients was middle aged or elderly, the last being a child of 8 months. A clinical summary of these cases is provided in Table 2. Significant risk factors for death included complicated disease (OR 30.9, 95% CI 3.4–276.6, $p<$ 0.0001), age $>$ 60 years (OR 8.8, 95% CI 1.7–44.9, $p=$ 0.006) and DM (OR 8.3, 95% CI 1.4–48.8, $p=$ 0.013) (Table 3). Logistic regression modeling identified complicated disease and DM as independent risk factors.

Complicated disease occurred in 21 patients(22.3%). The development of complications was significantly associated with the presence of SCD (OR 36.5, 95% CI 4.2–320.3, $p<$ 0.0001), invasive disease (OR 5.6, 95% CI 1.4–22.6, $p=$ 0.016), co-morbid illness (OR 3.5, 95% CI 1.3–9.5, $p=$ 0.011) and sepsis (OR 3.5, 95% CI 1.2–10.4, $p=$ 0.021). Logistic regression modeling identified SCD and sepsis as independent risk factors.

Mean LOH was 9.1 days (range 0–56) and was significantly increased 2 to 3-fold in patients with complicated disease, sepsis, CNS infections and isolates resistant to ampicillin or amoxicillin-clavulanate. Patients with invasive disease or isolates resistant to ceftriaxone had an over 4-fold increase in their mean LOH (Table 3). Multivariate linear regression identified ceftriaxone resistance ( $p<$ 0.0001), sepsis ( $p=$ 0.001), complicated disease ( $p=$ 0.008) and CNS infections ( $p=$ 0.013) as the only independent risk factors for an increased LOH.

4. Discussion

In our study we describe for the first time the morbidity and mortality of pneumococcal disease that is broadly representative of all age groups of the Jamaican population. The study population also comprised both hospitalized patients, who tend to have more severe disease, and outpatients, whose pneumococcal infection did not require hospitalization. A substantial portion of patients were within the paediatric age group ( $<$ 18 years of age), which is unsurprising given that children less than 5 years are the major reservoir for S. pneumoniae transmission [10, 17]. Adults over 60 years of age are also a group often affected by this infection given the waning immunocompetence that occurs at this age [15]. Our study identifies cases of pneumococcal disease in both these at-risk age groups and provides insight into aspects of their clinical presentation.

Our study highlights the unique challenges of pneumococcal disease management in children and adults in Jamaica and the region. Several risk factors for pneumococcal disease were present in a substantial proportion of our patients and several of these conditions pose serious public heath concerns. HIV/AIDS, for example, is a growing problem in the Caribbean, which is the second most affected area worldwide after Sub-Saharan Africa [7]. The most urbanized parishes in Jamaica report the highest prevalence of HIV cases $-$ 1570.1 cases per 100,000 persons in Kingston and St. Andrew and 2094.6 cases per 100,000 in St. James [13]. Patients with HIV infection have 10 to 100 times higher risk for pneumococcal bacteraemia and pneumonia and this association was reported previously in Jamaican patients with IPD [11, 15].

The prevalence of DM in the Jamaican population has been estimated at 8% and is of great concern for the adult population given the independent risk for mortality associated with DM in our pneumococcal disease cohort [14, 15].

Paediatric IPD in Jamaican SCD patients has been estimated at 480/100,000 individuals [6]. These rates are much lower than those reported in SCD patients in other developing countries with rates as high as 6900/ 100,000 [19]. Our relatively low rates are due in large part to the success of the Jamaican newborn screening program for SCD begun in 1995, which automatically enrolls SCD children in the Sickle Cell Unit (SCU). The SCU provides free ambulatory health care for these patients and the administration of monthly penicillin prophylaxis up to age 4 years and polysaccharide vaccine (PPV23) administration thereafter [6].

We report that SCD was most strongly associated with the development of complicated pneumococcal disease in this cohort of Jamaican patients, which in turn increases the risk for mortality. It is important to point out however that SCD itself was not associated with mortality in our study population. These factors underscore the need for PCV implementation in this patient population [1, 14].

Antibiotic resistance rates reported in our study are similar to previous estimates and indicate a relatively high resistance to beta-lactams and macrolides, a worrying global trend [16]. While the prevalence of resistance to ceftriaxone, ampicillin and amoxicillin-clavulanate was relatively low in our population, patients with isolates that were resistant to these agents had increased morbidity as measured by length of hospitalization. This did not, however, translate into an increased risk for the development of complicated disease or mortality. Detailed population studies are necessary to evaluate molecular determinants of resistance, transformation and competence in the evolution of S. pneumoniae infections across the country.

Our study is limited by the relatively small number of pneumococcal cases for review, however, we were able to show a strong association between complicated disease and SCD patients, which indicates that routine childhood pneumococcal vaccination in Jamaica may be beneficial. Studies investigating the island-wide prevalence of pneumococcal nasopharyngeal colonization in Jamaica, particularly in children, are urgently needed to clarify this point. In contrast, targeted use of vaccines in elderly patients with conditions such as DM, HIV infection or chronic respiratory diseases, which increase the risk of complicated pneumococcal disease, may be more ideal than routine administration given reported herd-protection effects [4]. Our findings will inform future efforts to improve the management of patients with pneumococcal infections and guide vaccine implementation strategies in Jamaica and the region.

Footnotes

Acknowledgments

The authors would like to thank the staff of the Microbiology Department at the University Hospital of the West Indies for their generous assistance in facilitating collection and processing of pneumococcal isolates at the institution. This research received no specific grant from any funding agency in the public, commercial, or not-for-profit sectors.

Conflict of interest

The authors declare that there are no conflicts of interest.

References

Allen

U.D.

Thomas

Carapetis

Henry

Wasfy

Lovgren

Richardson

and Low

D.E.

, Serotypes of respiratory tract isolates of Streptococcus pneumoniae from Jamaican children, Int J Infect Dis 7 (2003), 29–35.

Charalambous

B.M.

Batt

S.L.

Peek

A.C.

Mwerinde

Sam

and Gillespie

S.H.

, Quantitative validation of media for transportation and storage of Streptococcus pneumoniae, J Clin Microbiol 41 (2003), 5551–5556.

Cherazard

Epstein

Doan

T.L.

Salim

Bharti

and Smith

M.A.

, Antimicrobial Resistant Streptococcus pneumoniae: Prevalence, Mechanisms, and Clinical Implications, Am J Ther 24 (2017), e361–e369.

Corcoran

Vickers

Mereckiene

Murchan

Cotter

Fitzgerald

McElligott

Cafferkey

O’Flanagan

Cunney

and Humphreys

, The epidemiology of invasive pneumococcal disease in older adults in the post-PCV era, Has there been a herd effect? Epidemiol Infect 145 (2017), 2390–2399.

Feldman

and Anderson

, Recent advances in our understanding of Streptococcus pneumoniae infection, F1000Prime Rep 6 (2014), 82.

Hardie

King

Fraser

and Reid

, Prevalence of pneumococcal polysaccharide vaccine administration and incidence of invasive pneumococcal disease in children in Jamaica aged over 4 years with sickle cell disease diagnosed by newborn screening, Ann Trop Paediatr 29 (2009), 197–202.

Iannella

H.A.

and Luna

C.M.

, Community-Acquired Pneumonia in Latin America, Semin Respir Crit Care Med 37 (2016), 868–875.

Johnson

A.P.

Waight

Andrews

Pebody

George

R.C.

and Miller

, Morbidity and mortality of pneumococcal meningitis and serotypes of causative strains prior to introduction of the 7-valent conjugant pneumococcal vaccine in England, J Infect 55 (2007), 394–399.

Klemets

Lyytikainen

Ruutu

Ollgren

and Nuorti

J.P.

, Invasive pneumococcal infections among persons with and without underlying medical conditions: Implications for prevention strategies, BMC Infect Dis 8 (2008), 96.

10.

Lynch

J.P.

, 3rd and Zhanel

G.G.

, Streptococcus pneumoniae: Epidemiology, risk factors, and strategies for prevention, Semin Respir Crit Care Med 30 (2009), 189–209.

11.

McGregor

Barton

Thomas

and Christie

C.D.

, Invasive pneumococcal disease in Jamaican children, Ann Trop Paediatr 24 (2004), 33–40.

12.

Mitchell

A.M.

and Mitchell

T.J.

, Streptococcus pneumoniae: Virulence factors and variation, Clin Microbiol Infect 16 (2010), 411–418.

13.

MOH, HIV Epidemic Update - Facts and Figures, in, National HIV/STI Program, Jamaica HIV/AIDS Epidemic Update, 2011.

14.

Munroe

, Pneumococcal and vaccine use: Jamaican experience, 2008.

15.

Ortqvist

Hedlund

and Kalin

, Streptococcus pneumoniae: Epidemiology, risk factors, and clinical features, Semin Respir Crit Care Med 26 (2005), 563–574.

16.

Reinert

R.R.

, The antimicrobial resistance profile of Streptococcus pneumoniae, Clin Microbiol Infect 15(Suppl 3) (2009), 7–11.

17.

SAGE, Pneumococcal vaccines: WHO position paper, Weekly Epidemiological Record 14 (2012), 129–144.

18.

Spellerberg

and Brandt

, Streptococcus, in: Manual of Clinical Microbiology Jorgensen

J.H.

Pfaller

Carroll

K.C.

Funke

Landry

M.L.

Richter

and Warnock

D.W.

, ed., ASM Press, Washington, 2015, pp. 383–402.

19.

Wong

W.Y.

Overturf

G.D.

and Powars

D.R.

, Infection caused by Streptococcus pneumoniae in children with sickle cell disease: Epidemiology, immunologic mechanisms, prophylaxis, and vaccination, Clin Infect Dis 14 (1992), 1124–1136.