Non-Vaccine Serotypes Now Dominate Invasive Pneumococcal Disease in Three South African Provinces
This summary presents key findings from a laboratory-based surveillance study of invasive pneumococcal disease (IPD) conducted by the GERMS-SA network across three South African provinces — the Free State, KwaZulu-Natal, and Mpumalanga — covering 2019 to 2024. It examines incidence, age-specific burden, clinical presentation, in-hospital fatality, and serotype distribution relative to current pneumococcal conjugate vaccines (PCV10 SII and PCV13).
Streptococcus pneumoniae remains a leading cause of invasive bacterial disease in South Africa, particularly among infants and individuals with underlying co-morbidities. Despite the introduction of pneumococcal conjugate vaccines into the national immunisation programme, IPD continues to impose a substantial clinical burden.
South Africa introduced PCV7 in 2009 and subsequently transitioned to PCV13. More recently, PCV10 manufactured by the Serum Institute of India (PCV10 SII) has been in use. These vaccines target a defined set of pneumococcal serotypes, and prior surveillance data from the country have documented significant reductions in vaccine-type IPD following vaccine introduction. However, this has been accompanied by a relative and, in some settings, absolute rise in disease caused by non-vaccine serotypes — a phenomenon known as serotype replacement.
The GERMS-SA (Group for Enteric, Respiratory and Meningeal disease Surveillance in South Africa) network has maintained laboratory-based IPD surveillance since 2003. This study extends the evidence base with prospective data from three provinces over a six-year period, providing the most recent picture of serotype distribution, age-specific incidence, and outcomes in this geographic area.
IPD cases were identified through passive laboratory-based surveillance, with enhanced clinical data collection at a subset of sentinel sites.
| Data Component | Description | Coverage |
|---|---|---|
| Laboratory SurveillanceGERMS-SA network | Laboratory-confirmed IPD cases: S. pneumoniae isolated from a normally sterile site (blood, CSF, or other) or detected by PCR in a culture-negative specimen. Cases reported to a central laboratory database (CRDM). | Free State, KwaZulu-Natal, Mpumalanga — 2019–2024 |
| Enhanced Surveillance Sites (ESS)Clinical data collection | A subset of sentinel hospitals providing clinical data including demographics, HIV status, vaccination history, clinical presentation, co-morbidities, and in-hospital outcome for 36% (620/1,710) of cases. | Clinical data available for 82% (509/620) of ESS cases |
| SerotypingReference laboratory | Serotyping performed on isolates and PCR-positive, culture-negative specimens at the NICD reference laboratory. Serotypes classified against PCV10 SII and PCV13 vaccine coverage. | Serotyping data available for 50% of all cases (856/1,710); 63 cases not successfully serotyped |
Findings are presented across four domains: incidence and age distribution; clinical presentation and co-morbidities; in-hospital fatality; and serotype distribution.
2.3, 1.3, and 1.2 per 100,000 persons
Average annual IPD incidence was 2.3 per 100,000 in the Free State, 1.3 per 100,000 in KwaZulu-Natal, and 1.2 per 100,000 in Mpumalanga. Rates dipped across all provinces in 2020, likely influenced by COVID-19-related reductions in healthcare utilisation and transmission, before recovering from 2021 onwards.
Infants <1 year and adults 45–64 years
Infants under one year had the highest average annual incidence rates across all provinces: FS: 13.3; KZN: 10.2; MP: 9.2 per 100,000. A secondary peak was seen in adults aged 45–64 years (FS: 3.3; KZN: 1.7; MP: 1.6 per 100,000), consistent with the burden of underlying co-morbidities in this age group.
| Province | Dominant Presentation | Co-morbidities (excl. HIV) | ESS Cases with Clinical Data |
|---|---|---|---|
| Free State | Meningitis: 54% (76/141) | 40% (56/141) | 141 cases |
| KwaZulu-Natal | Pneumonia: 49% (124/253) | 37% (93/296) | 253–296 cases |
| Mpumalanga | Pneumonia: 45% (52/115) | 35% (40/115) | 115 cases |
25–40% across provinces
Where outcome was known (99%, 507/509 of ESS cases), in-hospital fatality was 40% in the Free State, 25% in KwaZulu-Natal, and 38% in Mpumalanga. Fatality was substantially higher in patients aged ≥5 years compared to children under 5: KZN 43% vs. 13% (81/190 vs. 8/62); MP 44% vs. 23% (37/84 vs. 7/24).
Bacteraemia without focus: highest in FS
In the Free State, bacteraemia without focal infection carried the highest fatality at 58% (11/19), followed by meningitis (48%, 36/75) and pneumonia (27%, 10/37). In KZN and MP, meningitis had the highest fatality rates: KZN: 45% (36/80); MP: 53% (19/36).
| Province | Non-Vaccine Serotypes | Most Common Serotype | Second & Third Most Common |
|---|---|---|---|
| Free State N=246 serotyped | 73% (180/246) | Serotype 8: 13% (31/246) | Serotype 3: 9% (23/246) | 19F: 9% (21/246) |
| KwaZulu-Natal N=415 serotyped | 68% (282/415) | Serotype 8: 12% (48/415) | Serotype 19F: 12% (49/415) | Serotype 3: 8% (35/415) |
| Mpumalanga N=195 serotyped | 66% (129/195) | Serotype 8: 16% (31/195) | Serotype 3: 14% (27/195) | 19F: 8% (16/195) |
The five-year GERMS-SA dataset confirms that invasive pneumococcal disease continues to impose a significant clinical burden in three South African provinces, with a disease profile increasingly dominated by non-vaccine serotypes.
The predominance of non-vaccine serotypes — comprising 66–73% of serotyped IPD across all three provinces — is consistent with trends observed globally following sustained PCV use. These findings should be interpreted in the context of documented temporal reductions in vaccine-type IPD following PCV introduction in South Africa; the current picture reflects a post-vaccine era in which non-vaccine serotypes have filled the ecological space vacated by vaccine-type disease.
Serotype 8 was the most commonly detected serotype in all three provinces. Notably, serotype 8 is not covered by PCV10 SII or PCV13, and the 23-valent polysaccharide vaccine (PPV23) — which does cover serotype 8 — was not reported to have been used in any of the individuals aged ≥5 years in this dataset. This represents a potential missed opportunity for protection in high-risk adult populations.
The marked provincial difference in dominant clinical presentation — meningitis in the Free State versus pneumonia in KZN and Mpumalanga — may reflect differences in the patient populations attending Enhanced Surveillance Sites, the age distribution of cases, and referral patterns to tertiary facilities, rather than true biological differences in organism virulence.
In-hospital fatality rates of 25–40% are consistent with the invasive nature of the disease and the clinical complexity of the affected populations, who carry a high burden of HIV infection and other co-morbidities. The substantially higher fatality seen in patients aged ≥5 years compared to children under five likely reflects the concentration of severe underlying illness in adolescents and adults accessing healthcare at the ESS hospitals.
This study highlights several surveillance and programmatic priorities relevant to the management of pneumococcal disease in the post-PCV era in South Africa.
GERMS-SA Network Infrastructure
A functioning multi-provincial laboratory surveillance network with standardised case ascertainment, central serotyping, and enhanced clinical data collection at sentinel sites.
PCV Programme Coverage
South Africa's childhood immunisation schedule includes a pneumococcal conjugate vaccine, and temporal reductions in vaccine-type IPD have been documented since its introduction.
Adult Pneumococcal Vaccination
No PPV23 vaccination was reported among individuals aged ≥5 years in this cohort, despite PPV23's coverage of serotype 8 — the dominant serotype across all three provinces.
Programmatic & Clinical Priorities
Key actions arising from the five-year GERMS-SA surveillance data
- Findings should be interpreted in the context of prior research documenting temporal reductions in vaccine-type IPD following PCV introduction in South Africa.
- The predominance of non-vaccine serotypes across all provinces underscores the need for ongoing serotype surveillance to inform future vaccine policy decisions.
- The high burden of IPD in infants under one year warrants sustained attention to timely vaccine delivery and high immunisation coverage in this age group.
- The absence of PPV23 vaccination among adults at ESS with serotype 8 IPD represents a potential gap in current adult pneumococcal vaccination practice.
- HIV co-infection and other co-morbidities are prominent drivers of IPD severity and should inform risk stratification and preventive care pathways.
This five-year laboratory-based surveillance dataset provides the most comprehensive recent characterisation of invasive pneumococcal disease in three South African provinces, and offers important insights for vaccine policy in the post-PCV era.
Serotype replacement is the defining feature of the current IPD landscape
Non-vaccine serotypes accounted for 66–73% of serotyped IPD across the Free State, KwaZulu-Natal, and Mpumalanga. Serotype 8 — not covered by PCV10 SII or PCV13 — was the single most prevalent serotype in all three provinces. This serotype shift has outpaced the vaccine-type reductions previously documented and represents the central challenge for future vaccine policy.
Infants bear the highest incidence, but adults face the highest fatality
Incidence peaks in infants under one year (up to 13.3 per 100,000 in the Free State), while in-hospital fatality rates are substantially higher in patients aged five years and older. A vaccine and prevention strategy that addresses both ends of the age spectrum — including high-risk adults — is needed to reduce the overall IPD burden.
The case fatality burden remains unacceptably high
With in-hospital fatality rates of 25–40% across provinces, and rates exceeding 50% for meningitis in Mpumalanga, IPD continues to carry a heavy mortality burden in South Africa. This underscores the urgency of prevention through broader vaccine coverage, and early identification and treatment of invasive disease.
Sustained surveillance is irreplaceable as the vaccine landscape evolves
Future changes to the national immunisation schedule — whether adoption of higher-valence vaccines, revised dosing schedules, or expanded adult programmes — will require robust, ongoing serotype surveillance to assess impact and detect emerging shifts. The GERMS-SA network remains the essential infrastructure for generating this evidence in South Africa.
- Bishop K, Legare N, Archary M, Bhola P, Dawood H, Khoza N, Mahabeer P, Mvelase N, Naby F, Ngubane WZ, Potgieter S, Ramjathan P, Sookan L, Tebeila N, de Gouveia L, Walaza S, Cohen C, von Gottberg A, Quan V, Meiring S, for GERMS-SA. Laboratory-based surveillance of invasive pneumococcal disease in the Free State, KwaZulu-Natal, and Mpumalanga provinces, South Africa, 2019–2024. Public Health Bulletin South Africa. 2026;4(1):217–252. [DOI / URL to be confirmed]
