Pneumococcal lactational mastitis in a healthy woman

Abstract

We report a case of lactational mastitis complicated by Streptococcus pneumoniae bacteraemia in a breast feeding, healthy woman in her 20s. Numerous investigations showed that mastitis was the probable source of S. pneumoniae bacteraemia. While S. pneumoniae is known to cause non-lactational mastitis in patients with underlying diseases, such as systemic lupus erythematosus, reports of lactating mastitis in healthy individuals are scarce, with only six cases reported in the scientific literature published in English since 1995. Similar to previous reports, our patient had a good clinical course with antimicrobial therapy, and the infection was presumably transmitted from the asymptomatic child to the mother. Although the exact mechanisms that establish transmission from a child remain unclear, both host and pathogen factors, such as stagnant milk or bacterial virulence factors, are thought to play a key role. Caution should be exercised because serotypes not currently covered by pneumococcal vaccines are emerging.

Background

Streptococcus pneumoniae, a commensal and conditional Gram-positive pathogen, commonly causes community-acquired pneumonia, sinusitis, otitis media and meningitis. However, it rarely causes skin and soft tissue infections. Such cases are only reported in patients with underlying diseases, such as systemic lupus erythematosus, diabetes mellitus, plasma cell dyscrasia, alcoholism, chronic renal failure, liver cirrhosis, nephrotic syndrome, HIV infection, chronic lymphocytic leukaemia and rheumatoid arthritis.^1–3 In severe cases, the 30-day mortality rate of S. pneumoniae cellulitis complicated by bacteraemia is reportedly as high as 16%.³ We report a case of an otherwise healthy lactating woman with pneumococcal mastitis complicated by bacteraemia with a good treatment outcome. Case reports published since 1995 are also reviewed.

Case presentation

A woman in her 20s who was breast feeding her 18-month-old son presented to the emergency department with a 24-hour history of fever and left breast pain. She also reported appetite loss, chills, vomiting and dyspnoea. Her medical history included postpartum depression. She had undergone a caesarean section but had no history of breast surgery. A non-smoker and occasional drinker, she also had no history of illicit drug use. All family members were healthy, and her child had received the pneumococcal conjugate vaccine series (PCV-13). On examination, the patient was in haemodynamic shock—blood pressure was 100/59 mm Hg, pulse rate was 113 beats per minute, body temperature was 39.7°C, respiratory rate was 22 breaths per minute, and oxygen saturation was 98% in ambient air. The left breast was moderately swollen with redness over the medial side of the upper and lower quadrants and significant tenderness on palpation. No fluctuance or mass lesions were detected. The breast was free of skin lacerations, ulcers, fistula, foreign bodies and tattoos. No nipple exudate or bleeding was noted. The right breast was normal, and an examination of the chest and abdomen was unremarkable. She had neither neck stiffness nor sinus tenderness and no erythema of her pharynx or tonsils. Her heart sounds were regular with no harsh systolic murmur. She had no joint swelling or oedema in the limbs. She had no skin rash, including Osler nodes or Janeway lesions. She had left axillary lymphadenopathy; however, the neck or groin area was not affected. Thus, a provisional diagnosis of septic shock and left breast inflammation was made.

Investigations

Routine laboratory investigations showed the following: slightly elevated white cell count (9.9×10⁹ /L) with bandemia; haemoglobin, 129 g/L; platelet count, 288×10⁹ /L; blood urea nitrogen, 17 mg/dL; creatinine 0.68 mg/dL; alanine aminotransferase, 12 U/L and aspartate aminotransferase, 12 U/L. We conducted a fever workup. Two sets of blood cultures showed S. pneumoniae growth. This S. pneumoniae was later found susceptible to penicillin G but resistant to erythromycin (minimum inhibitory concentration of penicillin G=0.06, erythromycin >2). The serotype was confirmed as 15C. Molecular analysis using multilocus sequence typing (ST) confirmed the isolated ST as 199. Urinalysis was negative for bacteria and pyuria.

Breast ultrasound performed to investigate the cause of left breast inflammation revealed moderate skin thickening with a poorly defined area of altered echotexture of inflamed fat lobules and glandular parenchyma, consistent with mastitis. Abscesses, mass lesions, calcifications or foreign bodies were absent. To investigate the source of S. pneumoniae bacteraemia, we also conducted contrast-enhanced CT of the neck, chest, abdomen and pelvis. The results showed only left breast mastitis as a possible source of bacteraemia (figure 1); no other foci of infection were detected. Infectious endocarditis was a potential cause of bacteraemia; however, transthoracic echocardiography showed normal valves with no vegetations.

Figure 1.

Contrast-enhanced chest CT revealed skin thickening of the left breast with poorly defined regions of increased density (white arrow), consistent with the findings of mastitis.

S. pneumoniae bacteraemia in otherwise healthy patients is highly unusual. Further clinical history revealed no record of surgical splenectomy and no family history of immunodeficiency disorders.

An immunology workup revealed negative HIV antigen/antibody, with no antibody deficiency (IgG, 960 mg/dL; IgM, 56 mg/dL; IgA, 268 mg/dL) or complement deficiency (complement C3, 92 mg/dL; complement C4, 23 mg/dL). A throat swab culture of her child was not taken because he had no signs of respiratory infection.

Differential diagnosis

S. pneumoniae isolated from the blood culture confirmed the diagnosis of invasive pneumococcal disease (IPD). As a potential source of bacteraemia, infective endocarditis was one of the differential diagnoses. However, S. pneumoniae is not a typical causative organism for infective endocarditis, and there were no clinical findings of infective endocarditis on physical examination or echocardiogram. Blood culture was repeated to check for persistent bacteraemia, which is key to diagnosing infective endocarditis; these were later found to be negative. Based on these findings, infective endocarditis was ruled out. Other differential diagnoses such as meningitis or sinusitis were unlikely considering her physical examination findings. The imaging studies also ruled out pneumonia or abscess formation as a source of bacteraemia, and there were no foci of infection other than left mastitis. Based on these findings, we concluded that lactational mastitis was the only potential source of bacteraemia. Other non-infectious differential diagnoses of mastitis, such as malignancy or granulomatous mastitis, were unlikely based on the rapid onset of presentation and findings from ultrasound. A non-infectious differential diagnosis of dyspnoea and shock, such as pulmonary embolism or heart failure, was ruled out based on the echocardiogram and CT scan findings.

Treatment

Empirical antimicrobial therapy with intravenous ampicillin/sulbactam was started before the blood culture results. She stopped nursing but continued bilateral breast pumping to enhance milk drainage.

Outcome and follow-up

Her condition improved significantly with antimicrobial therapy. She became afebrile, and her blood pressure normalised without vasopressor use. Her appetite loss and dyspnoea also improved, while the swelling and breast tenderness improved gradually. Based on the microbiological culture and sensitivity results, intravenous ampicillin/sulbactam was switched to intravenous penicillin G, and the leucocytosis normalised. She was discharged home with oral amoxicillin to complete 14 days of antimicrobial therapy. We encouraged her to discontinue nursing. At 1-year follow-up, she was still in good health.

Discussion

Lactational mastitis occurs in approximately 2%–10% of lactating women at any time during the lactation period but is most common during the first 2–3 weeks post partum, with 74%–95% of cases occurring by 12 weeks post partum.⁴ The causative organisms are coagulase-negative staphylococci, viridans streptococci, Staphylococcus aureus, Group B streptococci and Enterococcus faecalis.⁵ Septic shock secondary to lactational mastitis is infrequently reported, especially with S. aureus infections.⁶ Pneumococcal non-lactational mastitis has rarely been reported in patients with underlying diseases, such as systemic lupus erythematosus; however, reports of pneumococcal lactational mastitis are scarce, with only six cases reported in the English literature between 1995 and 2021 (table 1).^7–12

Table 1.

Reported cases of pneumococcal lactational mastitis

	Author/year	Country	Age	Immune status	Location	Temp WBC	Treatment	Clinical course	Child age	Child condition	Child vaccine	Milk culture	Blood culture	Culture from child	Reference
1	Wüst,7 1995	Switzerland	38	Immuno competent	Left upper and lower outer quadrants	38℃ 16 500	Flucloxacillin	Uncomplicated	9 mo	Rhinitis	–	Streptococcus pneumoniae 6B	–	S. pneumoniae 6B (culture from both nares and throat)	7
2	Kragsbjerg,8 1995	Sweden	38	Immuno competent	Left lower medial quadrant	39.4℃ 13 700	Cloxacillin ->benzylpenicillin ->phenoxymethylpenicillin	Uncomplicated	4 mo	Mild upper respiratory tract infection	–	S. pneumoniae	Negative	S. pneumoniae (culture from nasopharynx)	8
3	Appalaraju,9 2011	India	26	Immuno competent	Left lower quadrant	38.3℃	Amoxicillin/clavulanate ->linezolid	Breast abscess	–	–	–	S. pneumoniae (Resistant to penicillin)	Negative	–	9
4	Miedzybrodzki,10 2013	Canada	35	Immuno competent	Right lower lateral quadrant	39.0℃ 9800	Vancomycin and cefazolin ->cefadroxil	Uncomplicated	8 mo	Asymptomatic	PCV-7	S. pneumoniae 19A	Negative	Not performed	10
5	Hald,11 2018	Denmark	29	Immuno competent	Right medial side	39.7℃ 14 900	Dicloxacillin ->benzylpenicillin ->phenoxymethylpenicillin	Uncomplicated	3 mo	Asymptomatic	–	S. pneumoniae	S. pneumoniae 19F	Not performed	11
6	Skalidis,12 2019	Greece	37	Immuno competent	Left	40℃ 9200	Cefaclor	Uncomplicated	6 mo	Asymptomatic	PCV-13	S. pneumoniae 11	–	Not performed	12
7	Our case 2021	Japan	Late 20 s	Immuno competent	Left medial side	39.7℃ 9900	Ampicillin/sulbactam ->Penicillin G ->amoxicillin	Uncomplicated	15 mo	Asymptomatic	PCV-13	Not performed	S. pneumoniae 15C	Not performed	–

PCV-13, pneumococcal conjugate vaccine-13.

No regional differences existed in any of the reported cases. The mean age was 33 years, all seven cases (100%) had immunocompetent status, and both the right and left sides were affected. All seven cases (100%) had a fever over 38°C at the initial presentation, two of seven cases (29%)¹¹ presented with bacteraemia and were associated with septic shock. All seven cases (100%) had a good clinical course with prompt recovery after appropriate antimicrobial therapy. One case (14%)⁹ required incision and drainage for an abscess due to penicillin-resistant S. pneumoniae. Children of the affected mothers were less than 2 years old in all seven cases (100%). Except for two cases (29%)^{7 8} who had minor upper respiratory tract symptoms, all patients were asymptomatic. Paediatric pneumococcal vaccines were administered with PCV-7 in one case¹⁰ and PCV-13 in two cases.¹² The serotypes of S. pneumoniae detected in five cases^{7 10–12} were 6 B, 19A, 19F, 11 and 15C. In three cases,^{7 10 11} the infection could have been prevented by PCV-13 vaccination; two others¹² were not included in the PCV-13.

S. pneumoniae normally resides asymptomatically in the nasopharynx of children, with horizontal infection risk increasing in children under 2 years of age. Colony formation is age-dependent, with more than 50% colonisation by age 3 and a decline thereafter.¹³ Large cohort studies have reported that the prevalence of colonisation in children under 5 years of age is more than 90%¹⁴ compared with less than 5% in healthy adults.¹⁵ In our case, the source of S. pneumoniae bacteraemia was thought to be lactational mastitis. Considering the large difference in the prevalence of S. pneumoniae colonisation between adults and children, we assumed that the infection was established from the child’s nasopharynx via breast feeding. A throat swab was not taken from the child but the hypothesis was the infection had resulted from breastfeeding, and is supported by the findings of the other six cases described in the literature. Two cases^{7 8} showed that an infected child could cause S. pneumoniae mastitis in the mother. Four other cases^9–12 suggested an association with S. pneumoniae mastitis, even if the child was asymptomatic. The exact mechanism underlying this infection remains unclear; however, both host and pathogen factors are thought to play a key role.¹⁶ Host factors include stagnant milk, maternal stress, genetic factors, level of bactericidal molecules in milk and mammary-associated microbiota. Pathogen factors include the burden of bacteria and virulence factors.^16–18 These factors are thought to be intricately related to the pathogenesis of mastitis. In our case, the child was relatively big in terms of age; hence, the possibility that minor trauma from the child’s teeth could have induced infection cannot be ruled out, and we encouraged the mother to discontinue breast feeding.

IPD is defined as the isolation of S. pneumoniae from essentially sterile sites, such as the blood, spinal fluid and synovial fluid.¹⁹ Our case was considered to be IPD caused by S. pneumoniae serotype 15C (not covered by PCV-13), and genetic analysis revealed that it belonged to ST199. Since the introduction of pneumococcal vaccines, IPD caused by serotypes covered by pneumococcal vaccines has decreased significantly. In contrast, serotypes not covered by the vaccine have increased due to serotype replacement.¹⁹ In addition, through genetic transformation, S. pneumoniae can undergo serotype switching to acquire a different capsular polysaccharide synthesis locus.²⁰ Previously, the most important serotype replacement induced by PCV-7 was serotype 19A, which was also associated with penicillin and macrolide resistance, and a substantial proportion of those isolates belonged to ST199.²⁰ Following PCV-13 introduction, IPD caused by serotype 15B/15C belonging to ST199 has increased due to serotype switching,²⁰ and our case illustrates this phenomenon.

In conclusion, we report a case of serotype 15C pneumococcal lactational mastitis in an immunocompetent woman, presumably transmitted from her child. Even young, healthy women may develop IPD; however, they generally have a favourable course if treated with appropriate antimicrobial therapy. PCV-13 is essential for preventing IPD; however, as the number of serotypes not covered by PCV-13 increases, vigilance is required.

Learning points.

• Pneumococcal lactational mastitis is an extremely rare disease but can affect immunocompetent patients.

• Overall clinical outcome of invasive pneumococcal disease can be favourable if treated with appropriate antibiotics.

• Caution should be exercised because invasive pneumococcal diseases not covered by vaccines are emerging.

Ethics statements

Patient consent for publication

Consent obtained directly from patient(s).