Bacterial infections

Bacillary dysentery

Dysentery due to salmonella or shigella species tends to be self-limiting in immune-competent individuals. Because of some suggestions that individuals who did not receive antibiotics for salmonella infections may develop late salmonella osteitis, there is a movement that proposes treatment for all, preferably with a quinolone – e.g. ciprofloxacin 500 mg 2x/d for 5 days.

This is probably a reasonable option in the elderly or those with evidence of immune dysfunction (e.g. diabetes, on corticosteroids, or HIV) but there is little firm evidence to support the practice.

Cellulitis

Consider underlying causes – diabetes, mild cardiac failure, renal impairment.

Most cases are due to either streptococcal or staphylococcal infection, or a combination of the two. Cloxacillin 1g 6 hourly IV for 10 days (or flucloxacillin 500 mg 4x/d if mild) is active against both, so it is a logical single agent, unless there is reason to suspect resistance (e.g. if the patient has already been on a beta-lactam). In cases of nosocomial staphylococcal infection or penicillin allergy, consider clindamycin 300 mg 3x/d PO. It is well absorbed, so IV therapy is seldom needed – if it is, use clindamycin 600 mg 3x/d IV.

Duration of therapy is conventionally 7-10 days; although once there has been a satisfactory clinical response, intravenous therapy can be changed to flucloxacillin 500 mg 4x/d PO.

Necrotising fasciitis is a deeper infection with a wide range of possible pathogens including anaerobes, streptococci, staphylococci and gut organisms, and management is centred on full surgical debridement with broader-spectrum cover – e.g. cloxacillin plus gentamicin plus metronidazole.

Perspective – choice of route and agent in antibiotic treatment of cellulitis

The evidence is weak for any particular agent compared to another.¹ ,² A review drawing braver conclusions from the same shaky data found that covering for methicillin reistant staphylococci infection was not advantageous, and that co-trimoxazole was another agent worth considering.³ Another review looking at response times in uncomplicated cellulitis found that there should be a 30% reduction in the area of redness and in oedema by day 3.⁴ and that overall 12% fail to respond to initial antibiotic choice. With this sort of response rate, the lack of ability of small trials to pick up differences between antibiotics is not surprising; choices thus need to be guided by local resistance profiles.

Leprosy

The diagnosis is often missed, sometimes for years on end. Think about the possibility of this condition in anybody with anaesthetic skin patches, neuropathic symptoms, facial ‘thickening’ or a chronic blocked nose. The diagnosis is made by the triad of anaesthetic or hypoaesthetic skin lesions, thickened nerves, and finding of AFBs on skin scrapings, nasal smear or nerve biopsy. If organisms are found on scraping or smear, if there are more than five lesions or more than one nerve involved, then disease is regarded as multibacillary.

• Check the thickness of the ulnar nerve and posterior auricular nerves. If they are very easy to feel, then strongly consider the diagnosis. Look for hypoaesthetic areas, particularly on the back.
• Diagnosis is made by the characteristic clinical picture and the demonstration of AFBs.
• A nasal swab (often positive in florid lepromatous leprosy) involves nothing more complicated than inserting a pus swab in the patient’s anterior nares and sending it for AFBs. It sometimes makes the lab less confused if you smear the slide at the bedside.
• A skin scraping is performed by making a very superficial cut in the skin in a suspicious area (e.g. ear lobes) then squeezing the wound to evert it slightly, then with the back of the scalpel blade scrape down the edges of the wound and smear the material on a slide, air dry it and send to the laboratory to stain for AFBs.

Treatment of leprosy

This is with combination therapy and it is advisable to contact the local leprosy clinic. The key drugs are rifampicin and dapsone – paucibacillary disease can be treated with rifampicin 600 mg once a month and dapsone 100 mg daily for a total duration of therapy of 6 months. Multibacillary disease can be treated with the same rifampicin and dapsone regimen, but with the addition of clofazimine 50 mg daily, and with total therapy duration of one year.

The disease is curable.

Perspective – sensitivity of diagnostic signs in leprosy.

The constellation of findings of hypopigmented patches with loss of sensation, thickened peripheral nerves and AFBs on skin smears or biopsy had a sensitivity of 97% and a PPV of 98% in a study from Ethiopia.⁵ Enlargement of one more nerves is found in 76 – 96% of patients (less often in those with paucibacillary disease). Smears alone have relatively low sensitivity – about 30% in one report.⁶

Impetigo

Amoxicillin 500 mg 3x/d for 10 days is usually adequate therapy, remembering that many lesions looking like pure infections with strep pyogenes can also yield staph.

Pyomyositis

Deep abscesses in muscles, usually due to staphylococci, but sometimes streptococci, was originally recognised as a disease of the tropics. They are now seen in immunosuppressed individuals elsewhere as well: associations with HIV and autoimmune diseases such as SLE and RA, particularly when the latter two are on aggressive immunosuppressive therapy, are well described.

Patients present with pain and tenderness over the involved site, which can be anywhere where there is a muscle, but is quite commonly on the thighs. Because the abscess is deep, it is often not thought of initially, and alternative diagnoses include osteomyelitis (when there is associated fever and pain on walking), and DVTs. Extension of pus from another area is not strictly pyomyositis. The diagnosis may be overlooked for days or even weeks.

Once it is thought of, investigation depends on local facilities. Needle aspiration (18G or larger) is 100% specific but not all that sensitive. Yield can be increased by the recognition that thick pus may not be aspirable – even a ‘dry’ attempt should end with disconnection of the needle from the syringe, filling of the needle with air, and then reconnection with an attempt to blow the needle contents onto a slide. Sometimes a small but diagnostic bead of pus emerges.

If facilities for ultrasound or CT are available, these are far more sensitive and can be used instead of attempted blind aspiration. CT is more sensitive than ultrasound, but the latter is readily available and cheap. If it gives you the answer, there is little need to do a CT as well unless there is a suspicion that the US may have missed other pathology in a deeper compartment. If no facilities are available, the patient is toxic, and the diagnosis is seriously entertained, then surgical exploration should be considered even if aspiration is negative.

Definitive management is surgical, with antibiotics not considered an alternative, but a supplement to adequate drainage. Mortality should be very low if the condition is recognised and treated promptly.

Septic arthritis

This is usually due to staphylococcal or gonococcal infection. Suspect the latter in sexually active patients with associated tenosynovitis or skin rash and an oligo-arthritis. (In young patients also consider haemophilia with a joint bleed causing pain, swelling and warmth. In older patients think of gout.) Do a blood culture and aspirate the joint, then start therapy with ceftriaxone 1g daily IV and cloxacillin 2g 4x/d IV. If the culture show a gonococcus, one to two weeks of intravenous ceftriaxone is usually adequate, assuming a satisfactory clinical response. In penicillin allergic patients ciprofloxacin 500 mg 2x/d orally has been suggested. For staphylococcal infection treat for four weeks with intravenous cloxacillin, and consider surgical drainage if not settling promptly. For cloxacillin resistant staphylococcal infection consider vancomycin 1g 12 hourly.

Utility of diagnostic features in septic arthritis

A systematic review⁷ found that a synovial fluid WCC of > 50 000/mcL was present in 62% of patients with bacterial arthritis (spec 92%, LR+ 7.7, LR- 0.42.) . A synovial polymorph proportion of >90% was 73% sensitive (spec 0.79, LR+ 3.4, LR- 0.34.) Fever of >37.5 C (LR+ 0.67, and a synovial fluid protein of > 30 g/L (LR+ 0.90) were both present in less than 50% and had low specificities (0.31 and 0.46 respectively.)

Use of bedside urine dipstix has been a source of multiple articles of varying quality, but generally showing a sensitivity of 70 to 80% for diagnosing septic arthritis – e.g.⁸, with a similar specificity. Combining leukocyte esterase positivity with a negative dipstix glucose may increase sensitivity to 100% and specificity to 85%, with a useful positive likelihood ratio of 7.⁹

Septicaemia.

Even before the HIV era, septicaemia without an obvious source was well recognised, and in some series accounted for about a third of all cases. Late recognition of the condition, particularly in the elderly, may contribute to the rather poor outcomes often reported. Suspect it if any of the following are found:

• Unexplained hypotension or frank shock
• ‘Change of condition’ of a hospitalised patient previously progressing adequately
• New onset acidosis with no clear cause.
• Hypothermia.
• Either leukopenia or leukocytosis.

Management:

• Look hard for a cause.
• Fluid resuscitation if hypotensive or underperfused.
• Blood cultures.
• Start on a broad-spectrum antibiotic: e.g. ceftriaxone 1 g daily IV
• Antibiotic selection should be dictated by potential source as well as length of patient contact with the health system. Traditionally, when cover is not dictated by an obvious potential source, there is a tendency to use a beta-lactam combined with an aminoglycoside. This recommendation derives from the days when the beta-lactam was invariably penicillin. With greater gram-negative cover from the cephalosporins, the clinical benefit of the combination has recently been questioned.¹⁰ The studies on which this review was based were generally of poor quality, and because the prognosis is often poor in this condition, it is probably still quite reasonable to continue to use the combination until some better quality prospective data becomes available.

Perspective – gram negative septicaemia in HIV

One group of organisms, which are mentioned rather regularly in this context, are the non-typhi salmonella species, which can cause a variety of puzzling syndromes ranging from mild generalised debility to fulminant sickness. Management is as for any other gram-negative septicaemia, with attention to looking for a focus. In view of a high recurrence rate in HIV positive individuals, some authors recommend secondary prophylaxis with a quinolone for some months or until immunity has been restored with ART.

Syphilis serology.

• The management of patients with serum RPR (WR, VDRL) titres of more than 1:8 is relatively straightforward, and is available in most medical texts. However:
• The prevalence of false positive WRs (and VDRLs) increases with age and in some series is as high as 30% by age 70. These should be TPHA negative and of fairly low titre (e.g. 1:4 or less.)
• In the clinical context of syphilis being a possible cause of the patient’s problem, greater emphasis should be placed on low titres. Late HIV infection may also result in inability to mount a normal antibody response, and thus a low or negative result.
• The prozone phenomenon occurs in the presence of very high antibody levels – in this case there is a negative or very weakly positive result, which on dilution becomes obviously strongly positive.
• Low titre positivity may persist for years or even for life in the context of previously treated severe disease.
• If the clinical context does not favour syphilis as the aetiology (e.g. an incidental finding) and the titre is less than 1:8 and there is a history of recent treatment, then repeat the test after a month and treat if it is rising. If there is no history of previous treatment either give three weekly injections of benzathine penicillin or just follow up with a repeat RPR titre.

Perspective: neurosyphilis.

The CDC guidelines¹¹ for the diagnosis of neurosyphilis are:

• 1. ‘Positive’ WR (VDRL),
• 2. CSF WCC > 5 /mm³ and
• 3. CSF protein > 0.4g/l

Other more sensitive indices to consider in diagnosing neurosyphilis are:¹²

• 1. CSF-TPHA > 1:320, (specificity 100%, sensitivity 98.3%)
• 2. TPHA index > 70. (specificity 100%, sensitivity 98.3%)

(The TPHA index = CSF-TPHA/(albumin quotient x 1000), and the albumin quotient = CSF albumin/serum albumin (both expressed as g/l) i.e. if TPHA-CSF = 1:320, serum albumin = 40 g/l, and CSF albumin = 0.2 g/l, then albumin quotient is 0.005, and TPHA index = 64. Note that the CSF albumin must be requested – it is not the same as the total CSF protein!)

Treatment of neurosyphilis

Treatment for neurosyphilis is conventionally at least 20 megaunits of Penicillin G given in divided doses daily – e.g. 5 MU 6 hourly. In some ward situations actually getting 4 doses in in one day may be limited by nursing shortages. An alternative worth considering is ceftriaxone 2 g IV daily, although there is no credible comparative data comparing the two regimens. Treatment duration is listed as 10 to 14 days, and again with no controlled trial data to indicate which one is preferable. Generally, in a very well patient, it is probably reasonable to stop after 10 days, and in somebody who remains with symptoms, continue for the full 14 days.

Typhoid

This usually presents with fever and headache; the abdominal features may not be prominent – in particular there is often no diarrhoea. It is a form of gram negative septicaemia, so the presentation can be quite varied from an obvious enteric illness to someone with fever and psychosis, and almost anything in between.

Dysenteric presentations and presentations with cough and respiratory signs are described. If in doubt, do a blood culture and treat empirically as gram-negative sepsis. The relative bradycardia is rare – many patients have a tachycardia. The WCC is often not low, but tends to be normal rather than high unless intestinal perforation has occurred.

Treatment: ceftriaxone 2 g 2x/d IV for 10 days; can swith to oral ciprofloxacin 500 mg 2x/d when patient has improved, but ensure that the organism is sensitive before swopping. (Check your blood culture result!)

Remember to check stool cultures two and three weeks after treatment to ensure that a carrier state has not developed; if it has, treat with ciprofloxacin 5000 mg 2x/d for 6 weeks and do not allow food handling until 2 consecutive negative stool cultures are available.

Rickettsial infections.

Tick-bite fever

Infections due to Rickettsia africae resemble the textbook descriptions of scrub typhus – the term African tick-bite fever¹³ itself is still rarely used in textbooks. There is usually a period of about a week from the time of exposure until the clinical illness. People who lived in cities as children and hence were not exposed early are more likely to get the infection as adults. Descriptions of the disease in indigenous Africans are rare,¹⁴ perhaps reflecting childhood acquisition or under-reporting.

There is usually quite severe headache and varying degrees of fever. Most patients will have noticed the eschar, which can be multiple, is often inguinal and is usually associated with tender local adenopathy. A maculopapular or vesicular rash is reported in from 15 to 45%,¹⁵ but may be particularly difficult to see on darker skins. There may be a history of removing a red tick (Amblyomma) a few days earlier, and more rarely, the tick may still be attached. Very occasionally, there may be frank encephalitic or haemorrhagic features. Therapy is simple analgesia and doxycycline 100 mg twice daily for a week. In patients allergic to doxycycline, erythromycin or quinolones have been proposed, although efficacy data is scanty. In pregnancy clindamycin 600 mg 2x/d is considered a possibility.

---

1. Cross ELA, Jordan H, Godfrey R, et al. Route and duration of antibiotic therapy in acute cellulitis: A systematic review and meta-analysis of the effectiveness and harms of antibiotic treatment. J Infect. 2020 Oct;81(4):521-531. doi: 10.1016/j.jinf.2020.07.030. Epub 2020 Jul 31. PMID: 32745638. ↩

2. Mistry K, Sharma S, Patel M, et alJ. Clinical response to antibiotic regimens in lower limb cellulitis: a systematic review. Clin Exp Dermatol. 2021 Jan;46(1):42-49. doi: 10.1111/ced.14398. Epub 2020 Aug 29. PMID: 32860230. ↩

3. Brindle R, Williams OM, Barton E, Featherstone P. Assessment of Antibiotic Treatment of Cellulitis and Erysipelas: A Systematic Review and Meta-analysis. JAMA Dermatol. 2019;155(9):1033–1040. doi:10.1001/jamadermatol.2019.0884 ↩

4. Yadav K, Krzyzaniak N, Alexander C, et al. The impact of antibiotics on clinical response over time in uncomplicated cellulitis: a systematic review and meta-analysis. Infection. 2022 Aug;50(4):859-871. doi: 10.1007/s15010-022-01842-7. Epub 2022 May 20. PMID: 35593975. ↩

5. Britton WJ Lockwood DN. Leprosy. Lancet 2004;363:1209-19 ↩

6. Lockwood DN. Leprosy elimination: a virtual phenomenon or a reality? BMJ. 2002;324:1516-18 ↩

7. Margaretten ME, Kohlwes J, Moore D, et al. Does this adult patient have septic arthritis? JAMA. 2007;297:1478-88 ↩

8. Hassas Yeganeh, M., Talaei, M., Bazzaz, A.E. et al. Determination of diagnostic value (validity) leukocyte esterase (urine dipstick strip) in differentiating inflammatory arthritis from bacterial arthritis. Adv Rheumatol 60, 11 (2020). https://doi.org/10.1186/s42358-020-0115-3 ↩

9. Kolbeck L, Haertlé M, Graulich T, et al. Leukocyte Esterase and Glucose Reagent Test Can Rule in and Rule out Septic Arthritis. In Vivo. 2021 May-Jun;35(3):1625-1632. doi: 10.21873/invivo.12420. PMID: 33910845; PMCID: PMC8193298. ↩

10. Paul M, Benuri-Silbiger I, Soares-Weiser K, et al. Beta-lactam monotherapy versus beta-lactam-aminoglycoside combination therapy for sepsis in immunocompetent patients: systematic review and meta-analysis of randomised trials. BMJ. 2004;328:268-72 ↩

11. 1998 Guidelines for the Treatment of Sexually Transmitted Disease. US Department of Health and Human Services, Centres for Disease Control and Prevention, Division of STD Prevention, 23 Jan 1998, pp1-118 ↩

12. Luger AF, Schmidt BL, Kaulich M. Significance of laboratory findings for the diagnosis of neurosyphilis. Int J STD and AIDS 2000;11:224-234 ↩

13. Jensenius M, Fournier PE, Kelly P, et al. African tick bite fever. Lancet Infect Dis. 2003;3:557-64 ↩

14. Cohen GL, Blumberg LS, Karstaedt AS. Tick bite fever in black South Africans–a rare disease? J Infect. 1996;32:235-7 ↩

15. Raoult D, Fournier PE, Fenollar F, et al. Rickettsia africae, a tick-borne pathogen in travellers to sub-Saharan Africa. N Engl J Med. 2001;344:1504-10 ↩

Was this article helpful?

0 out Of 5 Stars

5 Stars		0%
4 Stars		0%
3 Stars		0%
2 Stars		0%
1 Stars		0%

5 Submit

Send Feedback

How can we improve this article?

Please submit the reason for your vote so that we can improve the article.

Submit