Background

Helicobacter cinaedi is a gram-negative spiral rod that was first discovered in the rectal culture from a man who had had sex with a man with proctitis [1]. H. cinaedi was thought to cause infection only in immunocompromised individuals; however, it has also been observed as a causative pathogen in immunocompetent patients [2,3,4]. H. cinaedi can cause bacteremia, skin and soft tissue infection, and arterial infection [5, 6].

Although infected (mycotic) aortic aneurysms are not common, they are difficult to treat and are associated with high morbidity and mortality. Mortality has been reported to be greater than 20%, usually attributable to delays in diagnosis and subsequent complications, such as rupture and sepsis [7,8,9]. Common pathogens include Staphylococcus aureus, Streptococcus pneumoniae, and non-typhoidal Salmonella followed by other gram-negative organisms such as Escherichia coli, Klebsiella, and Pseudomonas spp. [10, 11]. Mycobacterium spp., Treponema palladium, and Chlamydophila spp. have also been reported as causative pathogens, although rarely [12, 13].

Herein, we report three cases of aortic aneurysm infected by H. cinaedi that were successfully treated using meropenem plus surgical stent graft replacement or intravascular stenting. Furthermore, we performed a systematic review of the literature regarding aortic aneurysm infected by H. cinaedi.

Case series

Case 1

A 77-year-old immunocompetent man with a past medical history of hypertension and dyslipidemia presented to our department with a fever up to 38 °C and progressive left pleuritic chest pain for 2 weeks. On admission, the patient was not in acute distress and had a temperature of 37.6 °C, blood pressure of 143/76 mmHg, heart rate of 67/min, respiratory rate of 16/min, and oxygen saturation of 95% on room air. The patient was noted to have coarse crackles over the left lower lobe of the lung and tenderness of the lower abdomen without any rebound or guarding. Laboratory data showed a mildly elevated white blood cell (WBC) count of 8100/μL and C-reactive protein (CRP) of 15.0 mg/dL. Contrast-enhanced computed tomography (CT) chest-abdomen-pelvis examination demonstrated an aneurysm (30 × 42 mm) of the aortic arch with suggestion of a Stanford B dissection involving the descending aorta on a background of abdominal vessel wall thickening (Fig. 1), increased thickness of bilateral common iliac arteries (20 × 25 mm), and a 1 cm diameter low-density area in the spleen, compatible with infected vasculitis and splenic abscess.

Fig. 1
figure 1

Sagittal contrast-enhanced CT chest-abdomen-pelvis image demonstrates aorta wall thickening with pseudoaneurysm of the distal aortic arch and suggestion of Stanford type B dissection involving the descending aorta, with abnormal fluid extending to the level of the diaphragm

Full size image

We initiated ceftriaxone 2 g intravenously (IV) every 24 h, vancomycin IV 1 g every 12 h, and minocycline IV 100 mg every 12 h. Due to the active infection, we opted for conservative management without immediate surgery. Despite empirical antimicrobial therapy, the patient had progressive lower abdominal pain on hospital day 4, and a follow-up CT scan demonstrated enlargement of the aortic arch aneurysm (33 × 49 mm) and worsening aortic dissection. Therefore, total arch replacement was urgently performed on the same day. After surgery, the patient gradually improved. On hospital day 7, gram-negative spiral rods were cultured from blood samples obtained on admission (aerobic bottle, BacT/ALERT [bioMérieux, Inc., Durham, NC]). The empirical therapy was changed to meropenem 1 g IV every 6 h. Final blood culture results revealed H. cinaedi. The minimum inhibitory concentrations (MICs) measured by Etest (bioMérieux) for this strain were as follows: meropenem 0.008 μg/mL, penicillin G > 32 μg/mL, piperacillin/tazobactam 16 μg/mL, cefotaxime > 32 μg/mL, and levofloxacin > 32 μg/mL. Tissue culture of the infected aneurysm was negative. We continued meropenem for 6 weeks after surgery. Follow-up CT revealed resolution of the vasculitis involving the common iliac arteries without evidence of infection. As there are limited data on oral antimicrobials against H. cinaedi, we consulted the data for H. pylori. As faropenem was found to have good antimicrobial action against H. pylori in vitro [14], the patient was discharged on hospital day 46 with oral faropenem. He continued oral faropenem for 1 year, and his general status was stable at a 4-year follow-up.

Case 2

An 85-year-old woman with a past medical history of polymyalgia rheumatica treated with oral prednisolone 5 mg every other day and hypertension presented to our hospital for fever up to 38 °C and epigastric pain for 1 month that had not responded to a short course of oral cefcapene pivoxil. On admission, the patient was not in acute distress with a temperature of 36.9 °C, blood pressure of 158/48 mmHg, heart rate of 60/min, respiratory rate of 18/min, and oxygen saturation of 98% on room air. The patient was noted to have tenderness over the epigastric area but no rebound or guarding. Laboratory data showed an elevated WBC of 10,000/μL, (neutrophils 75%) and CRP of 5.27 mg/dL. Contrast-enhanced CT chest-abdomen-pelvis demonstrated increased wall thickness of the descending aorta (35 × 32 mm) and low attenuation surrounding the aorta, compatible with an infected aortic aneurysm.

We empirically initiated ceftriaxone 2 g IV every 24 h. Her fever subsided and general status gradually improved. On hospital day 5, gram-negative spiral rods were cultured from blood samples obtained at admission (aerobic bottle, BacT/ALERT [bioMérieux]) (Fig. 2). Empirical therapy was changed to meropenem 1 g IV every 6 h. On hospital day 7, 16S rRNA gene sequencing confirmed H. cinaedi. The MICs for this strain tested using the same method as for Case 1 were as follows: meropenem 0.008 μg/mL, ampicillin 32 μg/mL, cefotaxime > 32 μg/mL, and levofloxacin > 32 μg/mL. Follow-up CT on hospital day 17 revealed that the infected aneurysm had diminished (33 × 31 mm). Subsequently, meropenem was switched to oral faropenem on hospital day 44. Follow-up CT on hospital day 62 revealed enlargement of the infected aneurysm (35 × 42 mm). Thereafter, urgent intravascular stenting was performed. After this procedure, the patient improved and was discharged on day 76 without further complications. The patient continued oral faropenem and was clinically stable without any sign of recurrence at the 6-year follow-up.

Fig. 2
figure 2

Photomicrograph of Gram stain demonstrates gram-negative spiral rods, which grew on hospital day 5 from blood cultures taken at admission

Full size image

Case 3

A 72-year-old immunocompetent man with a past medical history of hypertension and benign prostate hypertrophy presented to the emergency room with difficult and painful urination for 10 days and lower abdominal pain for 3 days followed by fever up to 38 °C 1 day prior to admission. On admission, the patient was not in acute distress with a temperature of 37.8 °C, blood pressure of 120/70 mmHg, heart rate of 89/min, respiratory rate of 18/min, and oxygen saturation of 98% on room air. The patient was noted to have tenderness of the lower abdomen without any rebound or guarding. The prostate was elastic, soft, swollen, and tender. Laboratory data showed an elevated WBC of 9800/μL, (neutrophils 78%) and CRP of 18.9 mg/dL. Contrast-enhanced CT chest-abdomen-pelvis demonstrated increased wall thickness of the descending aorta with a wild, multilobulated appearance with focal outpouching (30 × 31 mm) and fat stranding surrounding the aorta (Fig. 3), compatible with an infected aortic aneurysm.

Fig. 3
figure 3

Coronal contrast-enhanced CT abdomen image demonstrates irregular wall thickening of the descending aorta having a wild, multilobulated appearance with surrounding soft tissue stranding

Full size image

We empirically started cefepime 1 g IV every 8 h, vancomycin IV 1 g every 12 h, and minocycline IV 100 mg every 12 h. On hospital day 12, the patient complained of worsening of lower abdominal pain, and a follow-up CT scan revealed enlargement of the aneurysm (36 × 35 mm). He underwent urgent abdominal aorta replacement on the same day. Tissue culture of the abdominal aneurysm was negative; however, 16S rRNA gene sequencing identified H. cinaedi from the tissue. We could not perform the susceptibility testing because H. cinaedi isolates could not be obtained from blood and tissue culture. The empirical therapy was changed to meropenem 1 g IV every 6 h. He was continued on meropenem for 4 weeks after surgical intervention and was discharged on hospital day 47 with oral faropenem. The patient continued oral faropenem for 3 years and was clinically stable without any sign of recurrence at the 5-year follow-up.

Systematic review

Two authors independently reviewed the titles and abstracts of database records, retrieved full texts for an eligibility assessment, and extracted data from these cases. A literature search was conducted in both the PubMed database (up to November 2019) using the keywords ((cinaedi) AND ((“Vascular Diseases”[Mesh]) OR (aneurysm*[TW]))) OR ((“Aneurysm, Infected”[Mesh]) AND (“Helicobacter”[Mesh] OR “Helicobacter Infections”[Mesh])) and the Embase database using the keywords (cinaedi OR ‘helicobacter cinaedi’/exp) AND (‘infected aneurysm’/exp. OR ‘aortic aneurysm’/exp. OR ‘aneurysm’) (Fig. 4). Knowing that there were several reports in Japanese papers, we have included those published only in Japanese in order to further understand the clinical characteristics of the disease by presenting more confirmed cases. In order to search for articles in Japanese, we used Ichushi, a major Japanese database, using the keywords ((cinaedi/AL)) or (“Helicobacter cinaedi”/TH)) and (((aneurysm [Japanese]/TH or aneurysm/AL)) or ((Vascular diseases [Japanese]/TH or Vascular diseases/AL)) or ((Artery [Japanese]/TH or Artery [Japanese]/AL))).

Fig. 4
figure 4

Flow chart depicts the systematic review process of this study

Full size image

We found ten papers describing thirteen cases of infected aneurysms due to H. cinaedi, all from Japan [6, 15,16,17,18,19,20,21,22,23]. Additionally, there were twenty-three confirmed cases presented in Japan at medical conferences, though not published in the peer reviewed literature. However, for these cases detailed patient information was not available. The clinical characteristics of the thirteen published cases, including our three cases, are shown in Table 1. Among them, the most common site of infection was the abdominal aorta (at least nine patients). Only two patient received oral corticosteroids; the others were immunocompetent. Regarding management, except for two cases with conservative management, the remaining eleven patients underwent in situ grafting, extra-anatomical bypass, or intravascular stent. Except one patient with fungal infection, all patients were successfully treated without further complications.

Table 1 Clinical characteristics of cases of aortic aneurysm infected with Helicobacter cinaedi
Full size table

Discussion and conclusions

This is the first systematic review of infected aortic aneurysms caused by H. cinaedi. The mechanisms underlying infected aortic aneurysms remain unknown. However, four main mechanisms have been hypothesized that result in infection of the arterial wall: 1) the development of infected aneurysms secondary to septic microemboli of the vasa vasorum, 2) extension from a contiguous infected focus, 3) hematogenous seeding of the intima during bacteremia originating from a distant infection, and 4) trauma to the arterial wall with contamination [24]. Recently, Araoka et al. reported that bacterial translocation from the intestinal tract could be a route that leads to H. cinaedi bacteremia [25]. The promotion of atherosclerosis by H. cinaedi has also been reported [26, 27]. In our three cases, none had evidence of infection with H. cinaedi at distal sites, extension from a contiguous infected focus, or a history of trauma; therefore, the development of infected aneurysms secondary to septic microemboli of the vasa vasorum or bacterial translocation from the intestinal tract were considered the most likely mechanisms.

There is a lack of data on susceptibility testing, and standard breakpoints of antimicrobial agents for H. cinaedi have not been defined. Previous studies reported that susceptibility testing for H. cinaedi isolates has been conducted using the agar dilution method [28, 29]. Low MIC values were generally reported for H. cinaedi strains for carbapenems, aminoglycosides, and tetracycline (MIC90 ≤ 1 μg/mL) [5, 30], moderate MIC values for ampicillin (MIC90 = 16 μg/mL), cefepime (MIC90 = 8 μg/mL), and ceftriaxone (MIC90 = 8 μg/mL) [5], and high MIC values for erythromycin (MIC90 > 64 μg/mL) [31]. Prior to 2000, low MICs of ciprofloxacin and macrolides were reported for H. cinaedi; however, the majority of H. cinaedi isolates have gained resistance since the early 2000s because of the increased use of these antimicrobials [30]. As the agar dilution method is not available in all institutes, the Etest could be an alternative method; however, it can be inaccurate because of the unclear edges around the growth inhibition zone [5]. Although there are no guidelines for recommended antimicrobial treatment, there has been varied used of beta-lactams such as penicillin, cephalosporin, and carbapenem [2, 32]. Monotherapy versus combination therapy has not been fully investigated. Further studies exploring the antimicrobial susceptibility profiles are warranted.

Except for one case, all of the reviewed cases underwent open repair (OR). OR is considered the gold standard with weak evidence supporting its superiority compared with endovascular aortic repair (EVAR). The optimal treatment of infected aneurysms remains unknown, although an increasing number of trials have reported EVAR as an alternative treatment for infected aortic aneurysms, with improved short-term survival compared with OR without the associated higher incidence of serious infection-related complications or reoperations [33]. In contrast, Luo et al. reported that persistent infection after EVAR does occur and is often fatal without surgical treatment [34]. The choice between OR or EVAR should be weighed against the risks and benefits for each patient.

The optimal duration of antimicrobial therapy for H. cinaedi infection also remains unknown. In addition to IV antimicrobials for at least 6 weeks after surgery, some experts recommend that patients continue oral antimicrobials after discharge for at least 6 months whereas others recommend lifelong therapy [8]. All of our patients continued antimicrobials for at least for 1 year.

Regarding the incidence and prevalence of H. cinaedi, all published reports of infected aneurysms caused by H. cinaedi are from Japan. The number of reports of H. cinaedi overall infection, including bacteremia and skin and soft tissue infection, has been increasing over the last 10 years, especially in Japan. It is thought that there are three main reasons for this. First, as H. cinaedi is becoming increasingly recognized in Japan, many institutions are extending the incubation period for blood cultures when clinicians suspect H. cinaedi infection. H. cinaedi usually grows slowly, and one clinical research study from Japan revealed that approximately 50% of H. cinaedi bacteremia would probably have been overlooked had the duration of monitored blood cultures been limited to 5 days [35]. After this report, many institutes extended the incubation period. Another hypothesis is that H. cinaedi may be related to Japanese-specific cultural behaviors such as eating raw fish as sushi and sashimi followed by colonization of H. cinaedi in the intestine. However, this is perhaps less likely given the increasing worldwide popularity of Japanese cuisine. Finally, some institutions have changed blood culture products, moving from BacT/ALERT to BACTEC (BD Diagnostics, Sparks, MD), and this could also be contributing to the increased detection of H. cinaedi [36]. New BacT/ALERT plus bottles that modify the antimicrobial-absorbing materials in the blood culture media could also have contributed to the improved detection [37, 38]. However, the true reason why H. cinaedi infections are increasing in Japan is as yet unknown. As these detections were made using the blood culture system described above for extended cultures and 16S rRNA gene sequencing in tissues, the sensitivity and specificity of the techniques for diagnosis and detection should be analyzed in the future. Furthermore, studies on the true mechanisms of H. cinaedi infections and its incidence and prevalence are warranted.

In summary, aortic aneurysms infected with H. cinaedi are rare and occur in immunocompetent as well as immunocompromised individuals. Clinicians should suspect H. cinaedi as a possible causative pathogen in patients with infected aortic aneurysms and extend the incubation period for blood cultures. The optimal choice of antimicrobials for H. cinaedi, the duration of therapy for infected aneurysms, and the optimal intervention regarding OR versus EVAR should be investigated in further studies.