The SGH laboratory has been doing antimicrobial surveillance testing on regular systematic sample collections of Neisseria gonorrhoeae sent to us from the Department of Sexually Transmitted Infection Control, National Skin Centre since 1992 (at least). These results are submitted to the World Health Organization Gonococcal Antimicrobial Surveillance Programme (WHO GASP). If you are interested, the reports are freely available at the WHO GASP website (https://www.who.int/reproductivehealth/topics/rtis/gonococcal_resistance/en/).
Ceftriaxone has remained a reliable antibiotic for a surprisingly long time but resistance has finally emerged. The first ceftriaxone-resistant N. gonorrhoeae was reported in Japan in 2011 (Ohnishi M, 2011). This strain named H041 had a ceftriaxone MIC of 2 mg/L. Since then, ceftriaxone-resistant N. gonorrhoeae have been reported in an increasing number of countries. Resistance is due to a mosaic penA allele (i.e. the gene is made up of bits of DNA from different genomes, N. gonorrhoeae can do this because they can acquire DNA from their environment by a process called transformation). This results in an altered penicillin-binding protein with reduced affinity to beta-lactam antibiotics.
Our lab diagnosed the first ceftriaxone-resistant N. gonorrhoeae to be isolated in Singapore in early 2018.
The isolate was resistant to ceftriaxone with an MIC of 1 mg/L (left) and an annular radius of 5 mm by the CDS method.
Whole genome sequencing was performed and analysis showed that the strain had the same penA allele as the FC428 clone. Furthermore it belonged to ST13871 (allele profile abcZ 126 adk 39 aro E67 fumC 987 gdh 148 pdhC 153 pgm 65) which is a single locus variant of ST1903 (allele profile abcZ 126 adk 39 aroE 67 fumC 157 gdh 148 pdhC 153 pgm 65), the common multi locus sequence type of the FC428 clone. More details can be found in our paper which has just been published (Ko K, 2019).
As many of the cases of ceftriaxone-resistant N. gonorrhoeae were reported to have been acquired in south-east asia, we can only expect this problem to become worse.
Lee K, Nakayama SI, Osawa K, Yoshida H, Arakawa S, Furubayashi KI, Kameoka H, Shimuta K, Kawahata T, Unemo M, Ohnishi M. Clonal expansion and spread of the ceftriaxone-resistant Neisseria gonorrhoeae strain FC428, identified in Japan in 2015, and closely related isolates. J Antimicrob Chemother. 2019 Apr 19. pii: dkz129. doi: 10.1093/jac/dkz129. [Epub ahead of print] PubMed PMID: 31002306. (no free access)
Ko KKK, Chio MT, Goh SS, Tan AL, Koh TH, Abdul Rahman NB. First Case of Ceftriaxone-Resistant Multidrug-Resistant Neisseria gonorrhoeae in Singapore. Antimicrob Agents Chemother. 2019 Mar 11. pii: AAC.02624-18. doi: 10.1128/AAC.02624-18. [Epub ahead of print] PubMed PMID: 30858209. (free access from May 2020)
It seems that every month there is a new article about a carbapenemase-producing Enterobacteriaceae (CPE) outbreak traced to sinks (please read about some local experience here and here). There is now even a sink laboratory! However there may be other reservoirs to consider in the hospital wet environment.
We recently were trying to trace the environmental source of a clone of CPE. Environmental swabs of the usual suspects-sinks, toilet bowl, toilet drains were all negative.
One tends to think of the sink and toilet first as potential reservoirs of CPE because patient secretions naturally find their way there. In retrospect, the shower drain is no less likely a place (especially if the patient is colonised or infected with CPE on the skin). In this case, the shower drain was the only environmental site sampled that was positive for CPE.
Perhaps shower drains may become the ‘next big thing’ when it comes to investigating CPE outbreaks.
Around the same time we were carrying out our investigation, an Australian team (Fernando SA et al, 2018) published a study (in a very similar clinical setting) where CPE was found to be heavily concentrated in the floor drains of the shower rooms (though that is not obvious from the title the paper). What is interesting about this article is that there was increased detection of CPE in the shower floor drains after cleaning as compared to before cleaning! This is similar to our own experience.
Now if you were to look down a typical shower drain you can see why. There is a lot of biofilm down there! This is not easy to remove completely and the cleaning process itself disrupts the biofilm resulting in greater release of CPE.
But isn’t the floor drain downstream of the patient? Is it just a case of a patient with CPE contaminating the floor drain? Like the situation with the sinks, we still need to think of a mechanism whereby CPE in the drain can somehow transfer to another hitherto uncolonised/uninfected patient.
A recent study has shown that using the shower produced aerosols containing carbapenemase-producing Pseudomonas aeruginosa from the floor drain (Hopman J et al, 2019).
In the long term this may require an engineering solution. A German group trying to control an outbreak of multidrug-resistant P. aeruginosa in a paediatric haematology unit redesigned their shower drains to be easily cleaned and disinfected and to reduce splashing (Kossow A, 2017).
Fernando SA, Phan T, Parker C, Cai T, Gottlieb T. Increased detection of carbapenemase-producing Enterobacteriaceae on post-clean sampling of a burns unit’s wet surfaces. J Hosp Infect. 2019 Feb;101(2):179-182. (no free access)
N. meningitidis is a Gram-negative diplococcus, The bacteria in these images don’t really demonstrate the kidney- or coffee- bean shape described in textbooks. The main clue that they are Neisseria spp. are the fact that the cocci occur in pairs.
Note that N. meningitidis sometimes resist the tendency to decolonize and may appear Gram-positive (Images 5 and 6).
Image 5b With ‘the eye of faith’ it is possible to discern a capsule (clearing of background material) around this apparently ‘Gram-positive’ diplococcus.
Image 6b There is a suggestion of capsule around these as well
Streptococcus minor is a recently described Gram positive coccus found in the tonsils and gastrointestinal tract of dogs, cats and a calf (Vancanneyt M, 2004). The most closely related species by 16S rRNA gene sequence similarity are the animal Streptococci-S. ovis (95.9%), S. gallinaceus (95.5%), S. iniae (95.4%) S. entericus (95%), and S. suis (94.1%).
Small colonies (hence minor) on sheep blood agar are surrounded by a narrow zone of alpha-haemolysis.
Gram stain of colonies growing on sheep blood agar.
Gram stain of colonies growing in broth culture.
The first case of human infection described was that of a 62 year old Dutch patient with peritoneal dialysis associated peritonitis (Dresselaars HF, 2014). The same strain of S. minor was isolated from the dialyse and the saliva of the patient’s cat.
The second case was a 51 year old Belgian woman who was bitten by a dog (Tré-Hardy M, 2016)
We did a retrospective review of all S. minor isolated in our laboratory from 2013 (when we started using MALDI-TOF for identification of bacteria) to 2017. The first isolate was in 2015 and there were a further 5 isolates from non-duplicate patients during the study period (3 each in 2015 and 2017). All patients had soft tissue infections involving pet bites (4 dogs and 2 cats).
This work was presented at the 28th European Congress of Clinical Microbiology and Infectious Diseases 2018. You can download and view the poster at the link below.
We are beginning an occasional series of blog posts on virology topics from our virology consultant Dr Wan Wei Yee.
Unfortunately Virology topics make for lousy blogging material as there are fewer opportunities for colorful and interesting pictures, but there is still scope for interesting discussions J. Apologies in advance for the wordy post…
Let’s discuss the investigation of OBI as I’ve observed many incomplete or inappropriate tests ordered prior to immunosuppressive treatment.
OBI is characterized by the persistence of Hepatitis B virus (HBV) genome in the liver of hepatitis B surface antigen (HBsAg) negative individuals, with/without detectable low level viraemia and with/without seropositive markers (e.g. anti-HBcore).
After acute HBV infection, the disappearance of HBsAg, presence of Hepatitis B core antibody (anti-HBcore) +/- Hepatitis B surface antibody (anti-HBs) indicates resolution. However, even after resolution, the virus lies dormant in the hepatocyte due to its ability to form a viral mini chromosome (cccDNA) within the hepatocyte nucleus which is kept in a latent state by the host’s immune system. That is until an opportunity arises leading to reactivation.
There is currently no curative and eradicative therapy available. Therefore, the purpose of screening for HBV prior to immunosuppressive treatment is to identify those who had previous exposure to HBV so that measures can be taken to prevent reactivation.
Why is this important? The prevalence of HBV infection in Singapore is intermediately endemic (5-6% in the 1980’s). Even though the incidence has decreased (3.6% in the 2010 seroprevalence study) as a result of the introduction of HBV vaccination in the childhood immunization program since 1987, nearly a quarter of the population have been exposed to HBV (22.5% anti-HBcore positive).1 Thus, OBI remains important in this substantial cohort due to it’s potential for reactivation as a result of increasing medical interventions which affects the immune system to treat numerous conditions in the aging population with increasing life expectancy.
Risk of reactivation is stratified into high, moderate or low risk depending on the HBV status and type of immunosuppressive treatment (refer to reference 2 for details).Based on this risk assessment, reactivation of HBV can be prevented with eitherprophylaxis or pre-emptive therapy (through close regular monitoring) with effective antivirals.
Therefore, as antivirals are potentially prescribed in OBI for certain immunosuppressive treatment to prevent reactivation, it is important to establish if indeed the patient has had true past HBV exposure.
Most would rely on screen positive anti-HBcore in the absence of sAg as indicator of OBI prior to immunosuppressive treatment. Beware that isolated core positive result can also be due to a false positive in the test assay (specificity of current assay is ≥ 98%) or if the patient had received blood products which could contain anti-HBcore.
It is important to review and look at previous results for evidence of HBV exposure (previous serological markers) and establish history of prior HBV vaccination. This will guide test selection and help with result interpretation.
Follow the algorithm for screening below for those who had not been screened recently and not vaccinated:
*Specificity of test assay is ≥98%, therefore there is a 2% incidence of false positive results.
1. Ang LW et al Epidemiology of hepatitis B virus infection in Singapore.. Epidemiological News Bulletin. 2010; 40 (4): 76-87
2. Loomba R, Liang TJ. Hepatitis B reactivation associated with immune suppressive and biological modifier therapies: current concepts, management strategies, and future directions..Gastroenterology 2017; 152: 1297-1309.
3. Gutiérrez-García ML et al Prevalence of occult hepatitis B virus infection.. World J Gastroenterol 2011; 17(2): 1538-1542.
4. EASL. EASL 2017 Clinical practice guidelines on the management of hepatitis B virus infection. J of Hepatology 2017; 67: 370-398.
We were recently sent a corneal specimen from a patient with suspected fungal keratitis.
On the KOH preparation of the specimen, broad hyphae with rare septa were seen (above and below).
Colony growth is typically white submerged colonies with few if any aerial hyphae (so it doesn’t look fluffy like a typical mould despite having hyphae).
5 days growth (front and reverse).
10 days growth (front and reverse). The radiate pattern of growth is most obvious in these images.
15 days growth (front and reverse).
Lactophenol cotton blue preparations of hyphae from colony growth.
Traditional culture-based identification of moulds is heavily dependent on observing the characteristic morphology of conidia (the asexual spores found on the tip of specialised hyphae).
So a lot of effort in the mycology lab is directed at getting moulds to sporulate and this used to account for a lot of the delay in lab diagnosis (you can imagine that it is not possible to identify a fungus by just looking at hyphae).
Because we were unable to get this isolate to sporulate, sequencing of the Internal Transcribed Spacer (ITS) was performed. This most closely matched that of Pythium insidiosum.
This is quite an important diagnosis to make.
Before the availability of ITS sequencing, the diagnosis of P. insidiousum was challenging and difficult, and it could also be under-reported because of lack of physician awareness (its still not in most microbiology tests). Knowledge of the clinical presentation with a high index of suspicion is required.
Though morphologically resembling fungi, P. insidiousum is not a true fungus and is more closely related to diatoms and algae. The cell wall is composed of a mix of cellulose-related compounds and glycan, and not chitin. Likewise, nuclei are diploid unlike fungi which are haploid.
This explains our difficulty with the morphological identification as formation of the characteristic oogonia (sexual stage) are rarely formed and zoospores requires special cultivation techniques using water agar and grass blades (Pythium spp. are usually plant pathogens).
Infections are associated with agricultural activity and are therefore rare in urbanised societies. However, they are well-described in the region particularly in Thailand and India (Pupabool J, 2006).
The most common infections are ocular as a result of direct inoculation. Vascular, cutaneous, subcutaneous and disseminated infections though very rare, have also been described. Vascular infections in particular are associated with thalassaemia and other haemaglobinopathies in Thailand.
Management is problematic as P. insidiousum does not respond to the usual anti-fungal treatment. Treatment options that have been attempted include surgery, itraconazole and terinfine, potassium iodide, antibiotics like azithromycin and linezolid, and immunotherapy with P. insidiousum antigen (Agarwal S, 2018).
Agarwal S, Iyer G, Srinivasan B, Benurwar S, Agarwal M, Narayanan N, Lakshmipathy M, Radhika N, Rajagopal R, Krishnakumar S, K LT. Clinical profile, risk factors and outcome of medical, surgical and adjunct interventions in patients with Pythium insidiosum keratitis. Br J Ophthalmol. 2018 Sep 11. pii: bjophthalmol-2017-311804. (no free access)
Have you ever noticed what appear to be tiny moths on the wall? These may not actually be moths (order Lepidoptera), but flies (order Diptera). They are commonly called drain, sink, filter, or sewer flies, which hints at their unsavoury affinity for plumbing and sewage systems,
If you were to google ‘drain fly’ you would discover a whole corpus of literature out there on how to get rid of them. Having experienced first hand a recent infestation in my shower, I can attest that it can be quite an annoying experience. This is because the flies are able to breed in shower drains or any cracks in the shower area that allow moisture to collect. For at least a week, drain fly larvae would mysteriously appear on the shower floor on a daily basis.
Drain flies on their own are of no medical significance. However the spontaneous appearance of the larvae in shower pans or in the toilet bowl have the potential to cause unnecessary alarm. Our laboratory occasionally receives larvae sent by patients who believe they have passed the worms in their urine or stool. However a careful history will usually reveal that the patient has found the larvae in the bathroom and assumed that it came from one of their body orifices.
There are a number of reports of urinary myiasis (infection with fly larvae) with Diptera in the literature, but the samples were invariably provided by the patient and the larvae could never be found on the patient during physical examination (see references below). Personally I would think it highly unlikely that this species could actually colonise the urinary tract without an exposed portal like a urinary catheter (and even then…). When the physician is interviewing the patient it is important to get a very clear history of how the worm or the ‘sample’ was obtained. It may well turn out that the ‘urine’ sample was obtained from the toilet bowl.
Once I found out more about the habits of drain flies, it occured to me that they have potential to be vectors of antibiotic resistant bacteria in hospitals. I was not the first one to think of this as the Germans seem to have done quite a bit of work in this area (Faulde M, 2013). The authors were able to isolate several potential nosocomial pathogens from the flies, though none of them were multi-drug resistant. I now look for drain flies when I go up to the hospital wards, but so far I have not seen any…
Faulde M, Spiesberger M. Role of the moth fly Clogmia albipunctata (Diptera: Psychodinae) as a mechanical vector of bacterial pathogens in German hospitals. J Hosp Infect. 2013 Jan;83(1):51-60. (no free access)
We recently isolated a Gram positive bacillus from the blood culture of a patient with urosepsis secondary to renal stones and hydroureteronephrosis.
Gram stain of film made from positive blood culture (anaerobic bottle). You can see the Gram stain appearance of P. lympophilum grown on solid agar in the articles by Williams GD, and Ikeda M.
Small grey colonies were seen on CDC anaerobe agar after incubation under anaerobic conditions.
The identity of P. lympophilum was suggested by MALDI-TOF (though the scores were low at 1.82 and 1.86), and confirmed by 16S rDNA sequencing.
Because of its anaerobic growth and formation of propionic acid, P. lympophilumused to be classified as a member of the genus Propionibacterium. However phylogenetic analysis of 16S rDNA sequences showed it did not cluster with Propionibacterspp. and there were differences in peptidoglycan composition. It has been therefore been given its own genus Propionimicrobiumgen. nov (Stackebrandt E, 2002).
As the species name suggests, the original strains were isolated from the lymph nodes of patients with Hodgkin’s disease (Torrey JC, 1916). It is now known to colonise the human skin and genital tract.
The first written case reports of infection (both were of the urinary tract) caused by P. lympophilum were only recently described in 2015 (Williams GD, 2015).
However, as pointed out by Williams GD et al, the Culture Collection of Gothenburg has 10 isolates of P. lympophilum in its collection dating back to 1991, including 3 isolated from blood culture and 2 from the urinary tract.
A more recent report by Japanese authors described P. lympophilum bacteraemia in a patient with difficulty urinating (Ikeda M, 2017).
Stackebrandt E, Schumann P, Schaal KP, Weiss N. Propionimicrobium gen. nov., a new genus to accommodate Propionibacterium lymphophilum (Torrey 1916) Johnson and Cummins 1972, 1057AL as Propionimicrobium lymphophilum comb. nov. Int J Syst Evol Microbiol. 2002 Nov;52(Pt 6):1925-7. (no free access)
Whipple’s Disease is a rare disease that classically is said to present with weight loss, diarrhoea, and joint pains but can affect multiple organs including, the nervous system, heart, lung, liver, kidneys, and skin.
When I was a medical student in the late 1980s. Whipple’s Disease was already thought to be due to an infection but the cause was still unknown. It was not until 1992, that the rod-shaped Tropheryma whipplei was identified as the bacterium responsible (Relman DA, 1992). This was a landmark paper because it was the first time a molecular method (16S ribosomal RNA sequencing) had been used to identify an infection caused by a bacterium that could not be cultured (at the time). For more information on 16S ribosomal RNA sequencing please see a previous post.
Since then, T. whipplei has been successfully cultured in cell lines (human embryonic lung) and even in axenic (without living material e.g. cells) media (Fenollar F, 2003). However because this type of culture is very specialized, lab confirmation of Whipple’s disease remains primarily by PCR.
We have had a number of specimens sent to us for diagnosis of Whipple’s Disease on the basis of a suggestive clinical presentation or a typical histopathological appearance (the underlying pathology is that of infiltration of tissues and organs with macrophages containing bacteria). However all have been negative by PCR up till recently.
On 22 June, our registrar Karrie received a request from the Infectious Disease physician for 16S rRNA PCR of a bioprosthetic heart valve. The patient had presented with valve failure and histopathology of the valve suggested endocarditis.
View of the explanted bioprosthetic valve from the bottom with vegetations seen prominently on the valve leaflet (Dr Timothy Tay, Department of Anatomical Pathology, SGH).
High magnification view of
the macrophages which display cytoplasm with a greyish hue (Dr Timothy Tay, Department of Anatomical Pathology, SGH).
Our cultures had so far been negative. We usually do not agree to 16S rRNA PCR direct from tissue sample unless organisms are visible (not in this case), however we proceeded on the basis of the histopathology report.
PCR of DNA extracted from the valve tissue was positive and the sequence matched that of T. whipplei.
You can copy the sequence above and submit it yourself to NCBI Nucleotide BLASTor the curated University of Hong Kong 16Spath DB 2.0. (click on BOLD to link to database websites)
We actually also have a T. whipplei specific PCR because broad-range
16S rRNA PCR is supposed to be less sensitive (von Herbay, A, 1996). It was of course a bit redundant in this case as we had already obtained the 16S rRNA gene sequence but nevertheless…
PAS diastase positive, coarsely granular cytoplasmic inclusions (representing aggregates of bacilli) present within the macrophages. Rhodococcus equi and nontuberculous Mycobacterium species can also give this appearance (Dr Timothy Tay, Department of Anatomical Pathology, SGH) .
The cytoplasmic inclusions also stain positive on GMS
(Dr Timothy Tay, Department of Anatomical Pathology, SGH)
Negative on Ziehl Neelsen stain
(Dr Timothy Tay, Department of Anatomical Pathology, SGH)
What’s next? Serendipitously, one of the pioneers of T. whipplei lab diagnosis, Dr Matthias Maiwald, is the microbiologist at our partner hospital the Kandang Kerbau Women’s and Children’s Hospital (Maiwald M, 1998, Strausbaugh LJ 2001, Bentley SD, 2003). Between us we will attempt to culture this organism. If we succeed, it may be the first isolation of T. whipplei from a clinical specimen in Singapore (or indeed in this part of Asia). Watch this space.
Renesto P, Crapoulet N, Ogata H, La Scola B, Vestris G, Claverie JM, Raoult D. Genome-based design of a cell-free culture medium for Tropheryma whipplei. Lancet. 2003 Aug 9;362(9382):447-9. (no free access)
von Herbay A, Ditton HJ, Maiwald M. Diagnostic application of a polymerase chain reaction assay for the Whipple’s disease bacterium to intestinal biopsies. Gastroenterology. 1996 Jun;110(6):1735-43. (no free access)
Bentley SD, Maiwald M, Murphy LD, Pallen MJ, Yeats CA, Dover LG, Norbertczak HT, Besra GS, Quail MA, Harris DE, von Herbay A, Goble A, Rutter S, Squares R, Squares S, Barrell BG, Parkhill J, Relman DA. Sequencing and analysis of the genome of the Whipple’s disease bacterium Tropheryma whipplei. Lancet. 2003 Feb 22;361(9358):637-44. (no free access)
Our Anatomical Pathology colleague was recently sent a stomach biopsy taken from an elderly patient who presented with an unrelated problem. He had a history of eosinophilia and eosinophilic colitis for a number of years though previous investigations for parasites were negative.
Slide stained with H&E scanned at 0.5X magnification. The area of interest is marked with an orange dot at left.
5X magnification with segments of larvae seen where indicated.
Since the patient had no travel history, it is possible he was infected many decades ago when Singaporeans were exposed to a less ‘sterile’ environment, and the parasites have since been maintained in his body by low level autoinfection in the gastrointestinal tract.