References

Berchtenbreiter C, Brüning R, Auernhammer A, Reiser M. Misleading diagnosis of retroperitoneal actinomycosis. Eur Radiol. 1999; 9:(9)1869-1872 https://doi.org/10.1007/s003300050937

Brook I. Actinomycosis: diagnosis and management. South Med J. 2008; 101:(10)1019-1023 https://doi.org/10.1097/SMJ.0b013e3181864c1f

Cintron JR, Del Pino A, Duarte B, Wood D. Abdominal actinomycosis. Dis Colon Rectum. 1996; 39:(1)105-108 https://doi.org/10.1007/BF02048278

Ferrari TCA, Couto CA, Murta-Oliveira C, Conceição SA, Silva RG. Actinomycosis of the colon: a rare form of presentation. Scand J Gastroenterol. 2000; 35:(1)108-109 https://doi.org/10.1080/003655200750024623

García-García A, Ramírez-Durán N, Sandoval-Trujillo H, Romero-Figueroa MDS. Pelvic actinomycosis. Can J Infect Dis Med Microbiol. 2017; 2017 https://doi.org/10.1155/2017/9428650

Harris LF, Kakani PR, Selah CE. Actinomycosis. Surgical aspects. Am Surg. 1985; 51:(5)262-264

Jung C, Wagner J, Heine B Successful outpatient management of pelvic actinomycosis by ceftriaxone: a report of 3 cases. Dtsch Med Wochenschr. 1999; 124:1066-1070 https://doi.org/10.1055/s-2007-1024483

Koren R, Dekel Y, Ramadan E, Veltman V, Dreznik Z. Periappendiceal actinomycosis mimicking malignancy report of a case. Pathol Res Pract. 2002; 198:(6)441-443 https://doi.org/10.1078/0344-0338-00279

Pérez-López F, Tobajas JJ, Chedraui P. Female pelvic actinomycosis and intrauterine contraceptive devices. Open Access J Contracept. 2010; 1:35-38 https://doi.org/10.2147/OAJC.S7476

Radu CC, Camarasan A, Podila CM, Perju-Dumbrava D. Sudden death of a teenager caused by Actinomycosis Israelii: a case report. Iran J Public Health. 2018; 47:(9)1413-1418

Read RC. Nocardiosis and actinomycosis. Medicine (Baltimore). 2005; 33:(5)114-115 https://doi.org/10.1383/medc.33.5.114.64953

Sánchez Hernández JAS, Merado Carrillo NA, Chilaca Rosas F, Rivera Tapia JA. Use of the IUD associated with secondary infection by Actinomyces in the female genital tract. Rev Esp Patol. 2004; 37:383-390

Spagnuolo PJ, Fransioli M. Intrauterine device-associated actinomycosis simulating pelvic malignancy. Am J Gastroenterol. 1981; 75:(2)144-147

Stringer MD, Cameron AE. Abdominal actinomycosis: a forgotten disease?. Br J Hosp Med. 1987; 38:(2)125-127

Sung HY, Lee IS, Kim SI Clinical features of abdominal actinomycosis: a 15-year experience of a single institute. J Korean Med Sci. 2011; 26:(7)932-937 https://doi.org/10.3346/jkms.2011.26.7.932

Valour F, Sénéchal A, Dupieux C Actinomycosis: etiology, clinical features, diagnosis, treatment, and management. Infect Drug Resist. 2014; 7:183-197 https://doi.org/10.2147/IDR.S39601

Wagenlehner FME, Mohren B, Naber KG, Männl HFK. Abdominal actinomycosis. Clin Microbiol Infect. 2003; 9:(8)881-885 https://doi.org/10.1046/j.1469-0691.2003.00653.x

Weese WC, Smith IM. A study of 57 cases of actinomycosis over a 36-year period. A diagnostic ‘failure’ with good prognosis after treatment. Arch Intern Med. 1975; 135:(12)1562-1568 https://doi.org/10.1001/archinte.1975.00330120040006

Westhoff C. IUDs and colonization or infection with Actinomyces. Contraception. 2007; 75:S48-S50 https://doi.org/10.1016/j.contraception.2007.01.006

Actinomycosis: a case report of two patients treated for the condition who required stoma formation

09 September 2021
Volume 30 · Issue 16

Abstract

Actinomycosis is a rare bacterial condition that is seen infrequently and no studies have examined its prevalence or incidence globally. The author, a stoma care clinical nurse specialist (CNS), found herself caring for two such patients 3 weeks apart, both of whom were diagnosed with pelvic actinomycosis. Both patients had been fitted with copper intrauterine devices (IUDs) 10 years previously and were not aware that leaving IUDs in situ for a prolonged period increased their risk of infection. This article gives an overview of the condition, with discussion around the two patients the author cared for within her own specialty; not all aspects of the condition are included as the she has no clinical experience in microbiology and is not a gynaecological CNS. The case studies show how both women presented and were treated, with interventions including emergency surgery, long-term antibiotic therapy and stoma formation. Pelvic actinomycosis is often associated with the use of an IUD, and greater awareness about the risk of leaving IUDs in situ for prolonged periods is needed among gastrointestinal, colorectal and gynaecological practitioners.

Actinomycosis is a rare, insidious disease (García-García et al, 2017). It occurs slowly (Wagenlehner et al, 2003) and subtly, but with harmful effects. To date, no epidemiological studies have been undertaken to determine its prevalence or incidence across the world (García-García et al, 2017). Centres and institutes globally appear to have encountered small numbers of patients with actinomycosis over many decades; clinical cases have been reported in Africa, Oceania, Asia, Europe and America. Some clinicians may never see it in their entire medical career (García-García et al, 2017).

Actinomycosis is a chronic bacterial infection caused by Actinomyces spp., which are Gram-positive, anaerobic bacteria (García-García et al, 2017), ie bacteria that do not require oxygen for growth. The bacteria spread by direct extension across tissue planes creating multiple abscesses, sinus and florid granulation tissue as they encroach on new tissue (Sung et al, 2011). This follows a breach in the normal mucosal barrier (García-García et al, 2017). Destruction of the mucosal barrier can occur through injury or trauma, surgery, inflammatory bowel disease or immunosuppression (Jung et al, 1999). Mucosal destruction allows the pathogen to cross the mucosal membrane or epithelial surface (Sánchez Hernández et al, 2004) and penetrate deep inside the tissue (Jung et al, 1999), where it proliferates and grows.

Actinomyces bacteria are opportunistic (Sánchez Hernández et al, 2004), affecting patients mostly when they are weak or immunocompromised. Actinomycosis is not contagious but it can be systemic, that is, it can affect the whole body (García-García et al, 2017).

Thirty species of Actinomyces have been identified, of which eight are associated with human infection (Brook, 2008). The most common is Actinomyces israelii.

Where the bacteria occur

The bacteria are found in multiple locations within and on the body. They have been found inhabiting the nose, mouth, throat, oropharynx, oesophagus, stomach and intestine. Other places they can occupy include the face, bone, skin, the central nervous system and the urogenital tract (García-García et al, 2017). Chiefly, they inhabit the oropharynx, gastrointestinal tract and urogenital tract (Valour et al, 2014). The perineum, where microorganisms can extend upwards to the cervicovaginal zone (from the anus), is another potential route of dissemination (Sánchez Hernández et al, 2004).

Ordinarily, the bacteria exist in normal flora (Stringer and Cameron, 1987), cohabiting harmoniously within organs without issue and without detection.

How is actinomycosis detected?

The bacteria are subjected to the Gram-staining process for preliminary identification. A violet dye is applied first, followed by a decolourising agent, then a second dye that stains red (García-García et al, 2017). Cell walls of some bacteria will retain only the first colour (violet), denoting that they are Gram-positive bacteria.

The process to detect the organism is undertaken histologically (Sung et al, 2011) and under anaerobic conditions (Spagnuolo and Fransioli, 1981). Seven days or more are required for the organism to grow and be cultured at 35°C (Spagnuolo and Fransioli, 1981; Wagenlehner et al, 2003).

Biochemically, leucocytes, C-reactive proteins (CRPs) (which are inflammatory markers), tumour markers (such as Ca-125 and alpha-fetoprotein) can be elevated in actinomycosis, with a deficiency in erythrocytes (erythropaenia) (Valour et al, 2014).

Diagnostic tools, such as MRI, CT, ultrasound and X-ray images and surgical laparoscopies are useful (Pérez-López et al, 2010), but do not provide a conclusive diagnosis (Berchtenbreiter et al, 1999). Surgery to obtain samples of tissue, fluid or pus is usually the preferred method when making a diagnosis (García-García et al, 2017).

Symptoms

Clinically, pelvic actinomycosis causes few symptoms. The initial presentation may include lower abdominal pain and fever, with or without a palpable mass (Sung et al, 2011). It can present with constipation and/or vaginal discharge (Valour et al, 2014).

Two patients reported feelings of extreme tiredness and exhaustion during discussions with the author. The duration of symptoms before diagnosis is usually 2 months (Valour et al, 2014), but for both of these patients this was not the case.

Abdominal actinomycosis and intrauterine devices

Actinomycosis of the urogenital tract is the second most common form of the disease (Valour et al, 2014), with pelvic actinomycosis being the principal presenting feature (Read, 2005). It can affect any age group, regardless of lifestyle or occupation (Sánchez Hernández et al, 2004).

Pelvic actinomycosis is often associated with the use of an intrauterine device (IUD) (Valour et al, 2014). Its symptoms can mimic those of malignancy of gynaecological origin such as uterine myoma or adenomyosis because of the presence of a solid abdominal component (Weese and Smith, 1975) which, when inspected surgically, is seen as a mass of fibrosed, dense tissue. Preoperatively, diagnosis is made in fewer than 10% of cases, because of its rare presentation and difficulty in confirming the pathogen (Harris et al, 1985). Diagnosis is usually made during laparotomy or laparoscopy (García-García et al, 2017).

IUDs can foster the growth of microorganisms, through wires left exposed in the exocervix (Cintron et al, 1996). Equally, changes to the metabolism of endometrial cells can further increase an inflammatory response (Sánchez Hernández et al, 2004).

García-García et al (2017) found that the literature showed that the greatest number of patients diagnosed with pelvic actinomycosis, on all continents, were using an IUD. Periods over which the disease was detected were in the range of 1–33 years. In the majority of cases, those with an IUD in situ for a prolonged period of time were more likely to be affected (García-García et al, 2017). Several authors recommend that an IUD is changed periodically to limit the development of actinomycosis, although no time frame has been stipulated. Valour et al (2014) recommend changing an IUD every 5 years at a minimum, Westhoff (2007) suggests every 5 years, while Sánchez Hernández et al (2004) recommend every 3 years.

Unfortunately for both presenting patients (Sandra Brown and Sarah Black—not their real names) their IUDs went unchecked for approximately 10 years and, during this time, bacterial colonisation was taking hold and having an effect.

Potential for mortality

The youngest known victim to have died from A. israelii was a 14-year-old boy living in Romania in 2016 (Radu et al, 2018). At autopsy, it was found that the bacteria had been aspirated from the oropharynx into his lungs and had subsequently spread to the heart, causing cell deterioration, myocardial infarction and death (Radu et al, 2018).

As stated above, A. israelii is the most common type of Actinomyces bacteria to cause human infection. The prevalence of death globally is not known, but one must accept that death can occur, particularly when the disease goes undetected, untreated and the burden of bacterial colonisation is great. Treatment, once infection is suspected or confirmed, is paramount.

Treatment

Antimicrobial therapy is the main treatment for pelvic actinomycosis (Valour et al, 2014). When the disease is caught early, it can be treated with antibiotics alone but, if disease is severe, surgery is needed.

Actinomyces is susceptible to most antibiotics but less so to metronidazole. The usual treatment consists of high and prolonged doses of intravenous penicillin G or amoxicillin for 4–6 weeks, followed by penicillin V orally for 6–12 months (Ferrari et al, 2000). In cases of penicillin intolerance, clindamycin, tetracycline and erythromycin can be used with good results (Sánchez Hernández et al, 2004).

The length of antibiotic therapy is dictated by the initial presenting infection, the development of any postoperative, intra-abdominal fluid collections and how long it takes the patient to recover.

It is also generally agreed (Ferrari et al, 2000; Koren et al, 2002; Valour et al, 2014) that a combination of complete surgical resection, followed by short-term antibiotic therapy is an effective intervention (Koren et al, 2002). In some instances when using a concomitant approach, antibiotic therapy can be reduced to 3 months (Valour et al, 2014).

Case studies

Before writing this article and presenting the clinical details of the two female patients, the author sought approval from each. Consent was granted by both women and, to ensure anonymity, their names have been changed. The confidentiality of each patient is maintained and upheld throughout the article.

Each patient has requested a copy of this article upon publication.

Case study 1

Sandra Brown is 49 years old, divorced with two children. She required emergency admission after a worsening 3-day history of abdominal pain. On admission, she was found to have sepsis and very high blood pressure, she was also dehydrated and had a stage 3 acute kidney injury.

She had felt unwell for 14 months. Four weeks before admission, Ms Brown had attended an outpatient gynaecological appointment for reported vaginal bleeding, weight loss and poor appetite. Actinomycosis was not considered as a differential diagnosis. She had had an IUD fitted 10 years before, and was waiting for an abdominal and vaginal ultrasound scan.

There were no clear lines of infection, despite stool cultures and blood cultures being taken (Table 1). Bloods were suggestive of malignancy. Ca19–9, an indicator for cancer, gave a result of 108 (normal range (0–34). Because of the deterioration in her condition, Ms Brown was taken to the intensive care unit (ICU) and, from here, she went straight to theatre. The surgical team were unsure of her diagnosis and had originally thought it to be tuberculosis or cancer.


Table 1. Sandra Brown's blood results on admission
Result Normal range
Haemoglobin (g/litre) 76 115–165
White blood count (x109/litre) 31.7 3.5–12
Albumin (g/litre) 18 35–50
Sodium (mmol/litre) 143 135–145
Potassium (mmol/litre) 5.0 3.5–5.1
Creatinine (μmol/litre) 113 49–92
eGFR (kidney function) (ml/min) 61 90–999
C–reactive protein (CRP) (mg/litre) 114 0–5
Magnesium (mmol/litre) 0.49 0.65–1.05
Oxygen saturations (%) 92 95–99
Platelets (x109/litre) 435 140–400

Operative procedure

A laparotomy was carried out, an omental biopsy taken, a loop ileostomy created and the IUD removed. Surgery was minimal, because the internal organs were severely inflamed and distorted. The abdomen was hostile and diagnosis was still uncertain. Samples of tissue and abdominal fluid were taken intraoperatively for analysis and histological assessment.

Postoperatively

Ms Brown was seriously unwell and was admitted to a level 3 critical care bed. She was ventilated for 4 weeks, but remained in the ICU for 5 weeks. She developed multiple intra-abdominal collections, for which numerous drains were inserted radiologically, and required dialysis when her kidneys failed.

The development of ascitic fluid intra-abdominally required aspiration (drainage). Ascitic fluid is a build-up of fluid within the peritoneal cavity greater than 25 ml in volume. A sample of this fluid was sent for histological assessment to establish whether any contained malignant cells, which would further support a diagnosis of cancer.

Because of weight loss before and during admission, Ms Brown required total parenteral nutrition to meet nutritional requirements. She developed critical care-associated myopathy so required extensive rehabilitation in an external facility after discharge and before going home.

Four years on, Ms Brown is yet to have her ileostomy reversed because of comorbidities and health issues relating to the longevity of the ileostomy. She no longer requires dialysis.

Case study 2

Sarah Black is 49 years old, married with two children. She required emergency admission after a worsening 2–day history of abdominal pain. Ms Black had felt unwell for 5 months. She had non-specific, vague symptoms, with diarrhoea and abdominal pain that had become debilitating. Ms Black had had an IUD fitted 10 years previously.

The team was uncertain of her diagnosis. There were no clear lines of infection, despite blood, stool and parasitic cultures. A test for tuberculosis proved negative. The blood results showed increased levels of inflammatory markers (CRPs) and white blood cells (Table 2). Ascites and a solitary liver lesion were noted on CT, which were suggestive of peritoneal malignancy.


Table 2. Sarah Black's blood results on admission
Result Normal range
Haemoglobin (g/litre) 72 115–165
White blood count (x109/litre) 50.1 3.5–12
Albumin (g/litre) 24 35–50
Sodium (mmol/litre) 132 135–145
Potassium (mmol/litre) 4.8 3.5–5.1
Creatinine (μmol/litre) 131 49–92
eGFR (kidney function) (ml/min) 39 90–999
C–reactive protein (CRP) (mg/litre) 495 0–5
Magnesium (mmol/litre) 0.68 0.65–1.05
Oxygen saturations (%) 83 95–99
Platelets (x109/litre) 777 140–400

Surgery was performed within hours of admission.

Operative procedure

An exploratory laparotomy was carried out, a right salpingectomy, a sigmoid colectomy and Hartmann's procedure were carried out, and the IUD removed.

Findings

The findings included: a pelvic abscess with extensive pus in the quadrants; an abscess involving the right ovary, right Fallopian tube, uterus, sigmoid colon and upper rectum; a perforation in the sigmoid colon; and a small amount of diverticular disease.

Postoperatively

Six days postoperatively, Ms Black developed a collection of fluid at the splenic flexure, which required radiological drainage and the insertion of a pigtail drain. This was in place for 14 days. She was in hospital for 1 month. She did not require rehabilitation or dialysis. Six months of oral penicillin V antibiotics were required once intravenous antibiotics had been discontinued.

The liver lesion disappeared independently. Ms Black did not have cancer; the lesion was related to intra-abdominal inflammation, so no further intervention was required. Reversal of the colostomy took place 8 months later. There were no further admissions. Ms Black is currently well.

Discussion

Although Ms Brown and Ms Black presented with different clinical features, there were commonalities between the two cases. For example, they both presented with intense abdominal pain, intense fatigue and loss of appetite (Box 1).

Box 1.Common features of actinomycosis in both patients

  • Abdominal pain
  • An increased risk of infection caused by intrauterine devices being left in place for a prolonged period
  • Increased bacterial growth and colonisation
  • Increased symptoms of feeling generally unwell
  • Extreme tiredness/lethargy
  • Raised Inflammatory markers (C–reactive protein)
  • Reduced appetite
  • Weight loss
  • Raised white blood counts
  • Deranged bloods (biochemistry)

Symptoms had begun to manifest some 5–14 months earlier. Ms Brown had experienced symptoms for longer, was the sicker upon admission and required the most intervention. Ms Black had recently lost a parent and thought her fatigue was related to increased external stressors (such as dealing with her father's funeral and sorting out the old family home while working in a 9–5 job), while Ms Brown had experienced a decline in her health over time. She described her symptoms as being ‘non-specific’ and ‘something she was unable to explain or put a finger on’. It was only when Ms Brown developed intense abdominal pain over a 3-day period and subsequently collapsed at home that she realised something was seriously wrong.

Results from bloods taken in the accident and emergency department (Tables 1 and 2) show Ms Brown and Ms Black's sodium and potassium levels were within the normal ranges, while their kidney function (eGFR) was significantly reduced (at 39 and 61 respectively). Ms Brown required dialysis because her kidney perfusion was poor.

Both had elevated levels of CRP—a protein made by the liver when tissue is inflamed. Ms Black's CRP was four times higher at 495 mg/litre than Ms Brown's at 114 mg/litre. The normal range of a CRP in an individual in good health is 0–5. Thus, in both patients, a raised CRP indicated inflammation and infection, although its origin initially was unknown. Following surgery and the commencement of intravenous antibiotics, both women's CRP levels began to reduce. However, it took an additional 4–6 weeks for these to normalise. Equally, both had a significantly elevated white blood count, with Ms Black's being four times higher (50.1 x 109/litre) than the higher end of normal (12x109/litre). This raised white blood count further substantiated the presence of infection.

Both Ms Brown and Ms Black presented with low albumin levels (of 18g/litre and 24g/litre respectively). Albumin is a protein produced by the liver, low levels of which can indicate malnutrition or liver or kidney disorders. Both women's albumin levels indicated poor nutritional health. Ms Brown was malnourished because she had not eaten properly for many months and hence the need for total parenteral nutrition—a liquid feed delivered to the body via a peripherally inserted central catheter that guarantees a number of calories, minerals and vitamins are delivered over a 24-hour period or until such time as a patient can eat independently and maintain their required daily intake.

Both women whose cases are discussed in this article felt that they had ‘been let down by the system’, ‘felt forgotten’ and genuinely ‘had no idea that this infection could ever have happened’ to them. Both had assumed their IUD ‘was OK, that it did not require follow-up, assessment or indeed removal within a particular time frame’. They had not received this vital information at the point the device had been fitted. Each said ‘that had they known the facts, [they] would have made the necessary appointment’ with their GP practice or family planning clinic for its removal. As it was, Ms Brown's and Ms Black's IUDs went unchecked for a decade, while potential problems were building up as a result of the extended wear time of the devices.

Both Ms Brown and Ms Black survived actinomycosis, although recovery has been extremely difficult and challenging for the former. Recollection of what happened still haunts her.

Ultimately, it was the tissue biopsies, fluid samples and histological analysis that confirmed diagnosis.

According to in-house microbiology colleagues of the author it is likely that erosion of the protective layer on the IUD and oxidisation of the copper would have precipitated the women's bacterial infection. Although the author was unable to find evidence from other sources to substantiate this as a cause, it is a plausible explanation that is worth noting.

Conclusion

Actinomycosis is a rare, long-term condition caused by anaerobic, Gram-positive bacteria that normally colonise in the mouth and the gastrointestinal and genitourethral tracts.

No substantive global data are available relating to its prevalence. No major international or collaborative institutional studies exist. Any information that is currently available has come from individual case studies from multiple institutes. Authors have amalgamated case studies and information from around the globe to produce a substantive number of related documents and articles.

This article has explored the plight of two women, both admitted as emergencies and both diagnosed with abdominal actinomycosis. For both women, it was concluded from histological sampling, analysis and surgical intervention that the extended length of time their IUDs had remained in situ had led to their acute hospital presentation. Both women had been fitted with a copper IUD. As the author has learnt from undertaking this article, there are many models of copper and plastic IUDs and intrauterine systems, with different life spans. The copper products generally have a wear time of 10 years, although there are models with around 5-year product life spans. No definitive time scale has been discerned from the literature. The author believes that checking the device is left to the woman wearing it, so she must be vigilant and be aware of its renewal/expiry date. By doing so, the potential risk of developing actinomycosis could be reduced.

Currently, when an IUD is fitted at a GP practice, gynaecological clinic or family planning centre, written information and additional instruction is given. Pocket-size cards are provided to women, containing the detail of their specific device (model) but not necessarily its expiry date. The author feels that perhaps some means of call back or telephone consultation might be useful every 2–3 years, and certainly if the IUD has a 10-year expiry date. However, this is not current practice. Following insertion there is no follow-up, unless the woman seeks advice or experiences symptoms, such as vaginal discharge or bleeding.

From a clinical perspective actinomycosis should always be considered (by the physician) as a differential diagnosis when an unusual mass or abscess presents on abdominal CT. Physicians must also be aware that actinomycosis may mimic the malignancy process at various anatomical sites. Bacterial cultures and pathology are essential for ensuring that an accurate diagnosis is made. Prolonged, high-dose antibiotic therapy is essential, with the duration being reduced in patients who have undergone surgical resection and removal of the infected tissue.

The author, having encountered actinomycosis first hand, is now aware of this condition and the potential dangers posed to women whose IUD may go unchecked, or whose IUDs remains in situ too long and certainly beyond the product life span.

The author has found this topic thought-provoking and plans to share this information with colleagues and nursing staff on both the colorectal and gynaecological units.

KEY POINTS

  • Actinomycosis has existed for many decades globally, but its prevalence remains unknown
  • Clinicians are only now beginning to share crucial information with each another, writing of their experiences and encounters with individual patients
  • The symptomology of actinomycosis can imitate that of some pelvic tumours, with radiological imaging serving only to supporting this diagnosis further
  • Signs and symptoms can be subtle, often taking months to develop before the patient becomes systemically unwell
  • The treatment and outcome of actinomycosis primarily depend on its severity at the time of presentation

CPD reflective questions

  • If a patient is admitted as an emergency admission, what would make you consider the possibility of actinomycosis?
  • Do you consider you have the knowledge to consider actinomycosis as a possible differential diagnosis when no other diagnosis can be made?
  • Think about how you could share knowledge of this condition with colleagues to increase awareness of the condition