Clinical Microbiology and Infection 31 (2025) 1225e1227

Contents lists available at ScienceDirect

Clinical Microbiology and Infection

journal homepage: www.clinicalmicrobiologyandinfection.com

Letter to the Editor

Broth microdilution susceptibility testing reveals underestimation of

resistance to penicillin and high levels of resistance to ampicillin of
Streptococcus pneumoniae clinical isolates in Sweden
Alexandros Petropoulos 1, 2, 3, *, Eva Morfeldt 4, Kerstin Nyre

n 4, Karin Blomqvist 1, 2, y,
1, 2, 4, y
Birgitta Henriques-Normark
1)
Department of Microbiology, Tumor and Cell Biology, Karolinska Institutet, Stockholm, Sweden
2)
Division of Clinical Microbiology, Karolinska University Hospital, Stockholm, Sweden
3)
Division of Infectious Diseases, Karolinska University Hospital, Stockholm, Sweden
4)
Department of Bacteriology, The Public Health Agency of Sweden, 171 82 Solna, Sweden

a r t i c l e i n f o

Article history: about the accuracy of the susceptibility data upon which treatment
Received 15 January 2025 guidelines have been based.
Received in revised form Here we performed broth microdilution (BMD) on 891 pneu-
18 March 2025 mococcal isolates with penicillin MIC 0.5 mg/L, that were referred
Accepted 25 March 2025
Available online 3 April 2025
to the Public Health Agency of Sweden (PHA) during the years 2015,
2017, and 2019 as part of the mandatory national surveillance
Editor: L Leibovici program. The method used to obtain the referred MIC values was
not reported to the PHA, but GT was widely used in Sweden before
the EUCAST warning was published. Resistance to other b-lactam
antibiotics was not reported to the PHA. We performed BMD ac-
cording to the EUCAST methodology, using 96-well custom plates,
prepared by ThermoFisher, containing antibiotic concentrations in
2-fold dilution steps: penicillin (0.008e8 mg/L), ampicillin
To the Editor, (0.008e8 mg/L), cefotaxime (0.008e4 mg/L), meropenem (0.008e4
mg/L), vancomycin (0.12e2 mg/L) with S. pneumoniae ATCC 49619
Streptococcus pneumoniae (pneumococcus) is a major human as the control strain [3]. Essential agreement (defined as MIC values
pathogen causing upper and lower respiratory tract infections as within one 2-fold dilution step) and categorical agreement (defined
well as invasive disease, such as sepsis and meningitis. The as MIC values classified in the same susceptibility category S-I-R)
cornerstone of treatment strategies against pneumococcal in- were calculated. Serotyping was performed on all the referred
fections consists of b-lactam antibiotics [1]. However, resistance to pneumococcal isolates using gel diffusion and Quellung reaction as
b-lactams has been increasing globally and accurate susceptibility described previously [4]. Approval by the Swedish ethical review
testing is therefore crucial not only for effectively treating indi- authority (2022-06926-01) was obtained.
vidual patients but also for the development of clinical treatment By retesting the isolates with BMD, we found that resistance to
guidelines. Gradient tests (GTs) have traditionally been used for penicillin (EUCAST non-meningitis breakpoint: MIC >2 mg/L) was
penicillin MIC determination of pneumococcal isolates, but in detected in 25.6% (n ¼ 228 of 891) of the included isolates. In
November 2019 European Committee on Antimicrobial Suscepti- contrast, the clinical laboratories had reported that 3.8% (n ¼ 33 of
bility Testing (EUCAST) published a warning against the use of 870) of the referred isolates had a MIC greater than 2 mg/L
penicillin GT because of the risk for underestimation of resistance (Fig. 1(a)). The essential agreement was 67.3% and the underesti-
around the resistance breakpoint (0.5e4 mg/L) [2]. However, the mation of penicillin MIC was 32% (reported penicillin MIC vs.
extent of the level of underestimation by GT is not known, and the penicillin MIC with BMD excluding isolates with imprecise MIC and
potential underestimation of penicillin resistance raises questions isolates with BMD MIC >8 mg/L, Fig. 1(a)). Most of the penicillin-
resistant isolates, according to BMD, were isolated from nasopha-
ryngeal swabs (n ¼ 128), followed by lower respiratory tract
* Corresponding author. Alexandros Petropoulos, Department of Microbiology,
Tumor and Cell Biology, Karolinska Institutet, Stockholm, Sweden. specimens (n ¼ 54), and blood (n ¼ 18). When excluding isolates
E-mail address: [email protected] (A. Petropoulos). from the upper respiratory tract samples, the remaining isolates
y
Equal contribution.

https://doi.org/10.1016/j.cmi.2025.03.019
1198-743X/© 2025 European Society of Clinical Microbiology and Infectious Diseases. Published by Elsevier Ltd. All rights are reserved, including those for text and data
mining, AI training, and similar technologies.
1226 Letter to the Editor / Clinical Microbiology and Infection 31 (2025) 1225e1227

Fig. 1. (a) Distribution of penicillin MIC values of Streptococcus pneumoniae isolates included in the study obtained by broth microdilution (BMD) compared with the reported
penicillin MIC values. Imprecise reported isolates were reported as “I”, “R” or “MIC >0.5 mg/L” thus the specific MIC value was missing. Green: Essential agreement defined as MIC
values within one 2-fold dilution step. Red: MIC values that differ two or more two-fold dilution steps. (b) MIC50 (mg/L), MIC90 (mg/L), and percentage of S-I-R (according to 2023
EUCAST non-meningitis breakpoints) based on broth microdilution susceptibility testing for various cell wall antibiotics on S. pneumoniae isolates included in the study. (c) Dis-
tribution of serotyping of S. pneumoniae isolates included in the study according to penicillin MIC (BMD) S-I-R categorization (2023 EUCAST non-meningitis breakpoints). Red: R,
Yellow: I, Green: S. (d) Distribution of serotyping of S. pneumoniae isolates included in the study according to ampicillin MIC (BMD) S-I-R categorization (2023 EUCAST non-
meningitis breakpoints). Red: R, Yellow: I, Green: S. (e) Distribution of serotyping of S. pneumoniae isolates included in the study according to cefotaxime MIC (BMD) S-I-R
categorization (2023 EUCAST non-meningitis breakpoints). Red: R, Yellow: I, Green: S. EUCAST, European Committee on Antimicrobial Susceptibility Testing.
 Letter to the Editor / Clinical Microbiology and Infection 31 (2025) 1225e1227 1227

(n ¼ 393) from the lower respiratory tract and sterile sites showed Author contributions
penicillin resistance in 24.2% (n ¼ 95 of 393) according to BMD. In
comparison, the clinical laboratories reported penicillin resistance A.P., E.M., K.B., and B.H.-N. designed the study. The susceptibility
in only 3% (n ¼ 12 of 393) of the same group of isolates. Even among testing was performed by A.P. and K.N. The results were analysed by
the limited number of invasive isolates included in the study A.P., E.M., K.B., and B.H.-N. A.P. wrote the first draft of the manu-
(n ¼ 56), resistance to penicillin was much higher with BMD script. All authors contributed to the writing and revising of the
compared with the reported MICs (39.6% vs. 4.2%). manuscript. K.B., and B.H.-N. provided equal contribution.
Resistance to ampicillin was detected in 80.7% (n ¼ 719 of 891).
All isolates were susceptible to meropenem and vancomycin, but in
Transparency declaration
contrast, resistance to cefotaxime was observed in 3.9% (n ¼ 35 of
891) of the isolates (Fig. 1(b)). Most of the cefotaxime-resistant
Potential conflict of interest
isolates were obtained from nasopharyngeal swabs (n ¼ 18), but
some were retrieved from lower respiratory tract specimens (n ¼ 8)
The authors declare that they have no conflicts of interest.
and blood (n ¼ 3).
To study the epidemiology of the pneumococcal isolates, we
analysed the serotyping distribution (Fig. 1(c-e)). The most preva- Financial report
lent serotypes found among the penicillin-resistant isolates
(n ¼ 228) were in descending order 19F (29.8%, n ¼ 68 of 228), 19A The work was supported by grants from the Torsten So€ derberg
(20.6%, n ¼ 47 of 228), 11A (7.8%, n ¼ 18 of 228), 23F (7.5%, n ¼ 17 of Foundation, the Swedish Foundation for Strategic research, the
228), and 14 (6.6%, n ¼ 15 of 228) (Fig. 1(c)). Among the isolates Swedish Research Council, Stockholm County Council, and the Knut
resistant to ampicillin (n ¼ 719), the most abundant serotypes were and Alice Wallenberg foundation.
non-typeable (18.4%, n ¼ 132 of 719), 19F (17.7%, n ¼ 127 of 719),
35B (12.1%, n ¼ 87 of 719), 19A (10.8%, n ¼ 78 of 719), and 11A (5.7%, References
n ¼ 41 of 719) (Fig. 1(d)). Among the most prevalent serotypes of
the penicillin-resistant isolates, serotypes 14, 19F, and 23F are [1] Metlay JP, Waterer GW, Long AC, Anzueto A, Brozek J, Crothers K, et al. Diag-
included in both PCV10, and PCV13 (the vaccines used in Sweden nosis and treatment of adults with community-acquired pneumonia. An official
clinical practice guideline of the American Thoracic Society and Infectious
during the study period), whereas serotype 19A is only present in Diseases Society of America. Am J Respir Crit Care Med 2019;200:e45e67.
PCV13 (not in PCV10), and 11A is absent in both PCV10 and PCV13. https://doi.org/10.1164/rccm.201908-1581ST.
In conclusion, pneumococcal resistance to antibiotics such as b- [2] European Committee on Antimicrobial Susceptibility Testing. Warning against
the use of gradient tests for benzylpenicillin MIC in Streptococcus pneumoniae:
lactams is higher than previously reported, threatening effective EUCAST. 2019. Available from: https://www.eucast.org/fileadmin/src/media/
treatment, and correct diagnostics and accurate MIC values are of PDFs/EUCAST_files/Warnings/Warnings_docs/Warning_-_gradient_for_benzyl_
great importance for the treatment of individual patients and for and_pnc_21nov2019b.pdf. [Accessed 15 January 2025].
[3] European Committee on Antimicrobial Susceptibility Testing. MIC determina-
the development of treatment guidelines. In this study, 891 isolates
tion of non-fatidious and fastidious organisms: EUCAST. Available from:. 2023.
with penicillin MIC 0.5 mg/L were included, and although only a https://www.eucast.org/ast_of_bacteria/mic_determination. [Accessed 15
few were invasive (n ¼ 56, 6.3%), the results strongly suggest that January 2025].
€stro
[4] Galanis I, Lindstrand A, Darenberg J, Browall S, Nannapaneni P, Sjo €m K, et al.
broth microdilution should be implemented in susceptibility
Effects of PCV7 and PCV13 on invasive pneumococcal disease and carriage in
testing algorithms to ensure accurate susceptibility data for Stockholm, Sweden. Eur Respir J 2016;47:1208e18. https://doi.org/10.1183/
S. pneumoniae infections. 13993003.01451-2015.