Abstract
Summary
Infection is a common complication of advanced diabetic foot disease, increasing the risk of acute admission and amputation. It is less well-known that foot ulceration and osteomyelitis may cause bacteraemia-associated hematogenous seeding and subsequent epidural abscess formation. Here we describe the case of a 57-year-old woman with known diabetic foot ulcer with underlying osteomyelitis admitted with backpain in the absence of trauma. Her condition deteriorated secondary to overwhelming sepsis. MRI of the spine confirmed spondylodiscitis and posterior epidural collection, not amenable to surgical intervention due to patient’s comorbidities and high surgical risk. Despite prolonged antibiotic therapy, the patient died following a hospital admission lasting 2.5 months. This case highlights the importance of regular contact with diabetes foot service for optimisation and prompt treatment of diabetic foot disease, which can be an underestimated potential source of remote site invasive systemic infection. Secondly, high clinical suspicion in admitting clinicians is imperative in ensuring timely diagnosis and early intervention to minimise fatal consequences.
Learning points:
-
Approximately 10% of patients with diabetes will develop a foot ulcer in their lifetime.
-
Spondylodiscitis (incorporating vertebral osteomyelitis, spondylitis and discitis) is a rare condition and diabetes is the most common predisposing risk factor.
-
Spondylodiscitis often presents with no other symptom other than back pain. Neurological or infective symptoms can be present or absent.
-
High clinical suspicion in clinicians is imperative in ensuring timely diagnosis and early intervention to minimise devastating consequences.
Background
Spondylodiscitis (vertebral/spinal osteomyelitis or septic discitis) is inflammation of the intervertebral disc spaces and adjacent vertebral bodies due to an infection. Spondylodiscitis may be aetiologically classified as pyogenic (bacterial), granulomatous (tuberculous or fungal) or parasitic (Echinococcosis). Pyogenic spondylodiscitis is predominantly monomicrobial, with 20–84% of all cases caused by Staphylococcus aureus (1), and Streptococcus and Enterococcus species accounting for 5–20% of cases. Pyogenic spondylodiscitis is often introduced from a distant infection site via the haematogenous route (1). Once bacteria reach the metaphyseal vascular arcades, the cartilaginous end-plates act as potential areas of inoculation while the release of bacterial proteolytic enzymes destroy the intervertebral discs (2). Uncontrolled spondylodiscitis infection can lead to the formation of paravertebral abscesses and result in cavitation and compression fractures or a spread into the spinal canal which can then lead to meningitis, formation of epidural abscess and neurological impairment (1) (Fig. 1)
Pathogenesis of haematological spread of foot infection in patients with diabetes mellitus.
Citation: Endocrinology, Diabetes & Metabolism Case Reports 2021, 1; 10.1530/EDM-20-0129
Early diagnosis of spondylodiscitis is often delayed as it tends to present with nonspecific manifestations. The most common clinical presentation is back pain, often accompanied with tenderness, paravertebral muscle spasm and restricted spinal range movement during physical examination (1). MRI is considered the most effective method to detect and evaluate spondylodiscitis. The most helpful laboratory tests are usually Erythrocyte sedimentation rate (ESR) and C-reactive protein (CRP); although nonspecific both are elevated in most cases (3). The standard of care is a conservative treatment consisting of appropriate antibiotics in combination with effective pain relief and non-pharmacological regimes such as immobilisation and physical therapy. Surgical intervention is indicated for spinal cord compression with progressive neurological deficits, for debridement and drainage of enlarged abscesses (4).
An important primary focus of infection for spondylodiscitis is diabetic foot ulceration in patients with type 2 diabetes (T2D), and can present as a late complication. Here we present the case of a female with known T2D who presented with back pain due to haematogenous complication of her diabetic foot disease.
Case presentation
A 57-year-old woman with poorly controlled T2D (HbA1c 148 mmol/mol) presented to the emergency department with lower back pain in the absence of trauma. She had a 4-month history of a right hallux apical neuropathic ulcer complicated by osteomyelitis. Patient also had a non-healing wound following left forefoot amputation 12 months ago for an infected left fifth toe ulcer which she had first developed 17 months prior to this admission (Fig. 2). During the period leading up to admission, patient had not engaged with the diabetes foot service and her concordance with antibiotic therapy was uncertain. Her past medical history included peripheral vascular disease, peripheral neuropathy, diabetic retinopathy, diabetic nephropathy, atrial fibrillation and hypertension. Her WHO performance status score was 2. Her medications on admission included Metformin MR 1 g twice-daily, Lantus 18 units at night, aspirin 75 mg once-daily, bisoprolol 7.5 mg once-daily, cholecalciferol 20 000 units once week, ramipril 5 mg once-daily and rivaroxaban 20 mg once-daily.
Flow diagram summarising progression of foot disease in a 57-year-old woman with type 2 diabetes. *Failed to attend multiple follow-up appointments. Numbers represent months pre-final admission.
Citation: Endocrinology, Diabetes & Metabolism Case Reports 2021, 1; 10.1530/EDM-20-0129
On admission, the patient was acutely confused, and septic. She reported lower back pain in the absence of any trauma. She denied a headache or any respiratory, urinary and gastrointestinal symptoms. On examination, she was acutely confused with a BP of 102/84 mmHg, heart rate of 110/min and temperature of 33.7°C. Her heart sounds were normal with no added murmurs and chest was clear on auscultation. Abdominal examination was unremarkable. Her residual left foot amputation wound site revealed soft tissue infection with capillary refill time (CRT) of 3 s and biphasic Doppler signals on both posterior tibial arteries (PTAs) suggestive of good vascular supply to the feet. The right hallux ulcer was dry with no signs of acute infection. Examination revealed lumbar vertebral tenderness and bilateral lower limb weakness and absent reflexes.
Investigation
She was in acute renal failure with profound metabolic acidosis and oliguria. Venous blood gas was as follows: pH 7.13, pCO2 4.2 kPa, lactate 14.6 mmol/L, bicarbonate 10.3 mmol/L, base excess −18.9 mmol/L. Blood results on admission are presented in Table 1. Rest of confusion screen was normal. Her chest radiograph and urine cultures were negative. CT scan of head showed no acute intracranial haemorrhage, collection or infarct. A FAST (focused assessment with sonography in trauma) emergency scan revealed a collapsible inferior vena cava (IVC) and poorly filled right ventricle (RV) – in keeping with significant volume depletion – and excluded abdominal free fluid or abdominal aortic aneurysm (AAA). CT abdomen (non-contrast) showed no convincing intra-abdominal collections or source of sepsis. The most recent duplex ultrasound of the arterial system of both lower limbs showed the common femoral (CFA), proximal profunda, superficial femoral (SFA), popliteal, anterior tibial (ATA), posterior tibial and peroneal arteries to be diffusely diseased with calcified plaques. There was no evidence of any haemodynamically significant stenosis in the right leg with biphasic spectral waveforms in the right lower limb. However, the left leg displayed 50–55% haemodynamically significant stenosis, one at mid SFA and another at mid popliteal artery. Moreover, the crural arteries had dense calcification with small lumens. The left lower limb distal anterior tibial artery had flow from peroneal artery. All spectral waveforms of the left leg appeared triphasic at the CFA, tri/bi to the knee with a strong monophasic signal below the knee. No acute revascularisation was deemed necessary based on these results. Wound swabs from the left forefoot amputation site had mixed growth including Pseudomonas sp. and yeast. Right foot wound swabs were negative for any growth. Previous foot ulcer swabs grew mixed anaerobes, S. aureus and Pseudomonas sp. Blood cultures grew Beta haemolytic group B Streptococcus (GBS) in the blood, suggestive of the diabetic foot ulcer as the origin. A transthoracic echocardiogram ruled out vegetations and demonstrated severely impaired systolic function (Visual EF of 25–30%). She was too unwell to undergo a transoesophageal echocardiogram. Abdominal imaging was unremarkable. Spinal MRI was arranged at this point due to it being the only objective source of pain and to exclude an underlying spinal abscess or discitis.
Blood results on admission (day 1), day 3 of admission during multi-organ failure (MOF) and terminal week.
Blood parameters | Reference range | On admission (day 1) |
MOF (day 3) | Terminal week (day 60) |
---|---|---|---|---|
Haemoglobin (g/L) | 114–150 | 99 | 92 | 92 |
Platelet (×109/L) | 135–400 | 172 | 121 | 339 |
White cell (×109/L) | 4.2–11.2 | 9.6 | 19.7 | 7.0 |
Neutrophil (×109/L) | 2.0–7.1 | 8.7 | 17.8 | 6.2 |
Lymphocyte (×109/L) | 1.1–3.6 | 0.6 | 0.8 | 0.4 |
Sodium (mmol/L) | 133–146 | 126 | 126 | 137 |
Potassium (mmol/L) | 3.5–5.3 | 4.7 | 4.3 | 4.7 |
Urea (mmol/L) | 2.5–7.8 | 18.4 | 26.4 | 7.4 |
Creatinine (mmol/L) | 55–110 | 241 | 233 | 60 |
Total bilirubin (µmol/L) | 0–21 | 25 | 24 | 3 |
ALT (IU/L) | 0–34 | 14 | 1034 | <6 |
AST (IU/L) | 0–40 | 56 | 2053 | – |
ALP (IU/L) | 30–130 | 338 | 371 | 173 |
GGT (IU/L) | <40 | 223 | 245 | – |
Total protein (g/L) | 60–80 | 52 | 51 | 58 |
Albumin (g/L) | 35–50 | 12 | 11 | 13 |
Globulin (g/L) | 19–35 | 40 | 40 | 45 |
Glucose (mmol/L) | 3.0–7.8 | 16.0 | 6.8 | 8 |
C-reactive protein (mg/L) | 286.4 | 154.1 | 17.3 |
Treatment
She immediately required inotropic support in the high dependency unit. She was treated for severe sepsis likely secondary to diabetes foot infection with IV Ceftriaxone and metronidazole. Her glycaemic control was optimised with a variable rate of insulin infusion. Vascular team reviewed the patient on admission and concurred with IV antibiotics for infected diabetes foot disease with no immediate indication for urgent debridement or surgery.
Outcome and follow-up
Despite maximum critical care support, she was in multi-organ failure with the combination of severe left ventricular dysfunction (LVF), acute kidney injury, and ischaemic liver dysfunction (Table 1). MRI of spine confirmed severe osteomyelitis with L3/L4 and L4/L5 discitis, and adjacent vertebral end-plate oedema. In addition, she had a posterior epidural collection with canal compression extending from T12 to L4 (Fig. 3). She was managed conservatively with prolonged courses of antibiotics (ciprofloxacin & gentamicin, then meropenem, and finally ceftriaxone) with ongoing microbiology and diabetes foot team input as neurosurgical intervention was deemed to carry a significant mortality risk and she declined the surgery.
MRI of spine T2 sagittal images. The green arrows point at L3/L4 and L4/L5 discitis with vertebral end-plate oedema. The red arrow points at the posterior epidural collection which caused canal compression extending from T12 to L4.
Citation: Endocrinology, Diabetes & Metabolism Case Reports 2021, 1; 10.1530/EDM-20-0129
Over the next several weeks of the patient’s inpatient care, inflammatory markers improved (Table 1) but remained bed bound with persistent lower limb neurology. In her final week she exhibited further signs of sepsis, despite prolonged antibiotic therapy and passed away 2.5 months into her admission. Her cause of death was recorded as (i) Septicaemia, (ii) Epidural abscess and (ii) Diabetic foot ulcer.
Discussion
Few case reports have previously described diabetic foot ulcers (DFU) as potential primary source of distant infections such as vertebral osteomyelitis (5, 6, 7). DFU are exposed to skin commensal bacteria that can colonise the wound as multi-layered microbial communities. These exist in structurally complex extracellular biofilm layers and confer both protection from the host immune system (e.g. low permeability to antibodies and neutrophils) and antibiotics. GBS has been reported as a prevalent microorganism in DFUs (8). However, the most common pathogen in DFU (with or without osteomyelitis) is S. aureus. It is also the most common pathogen in vertebral osteomyelitis (2). A poor glycaemic control can impair immune responses and together with a lack of regular wound surveillance favour wound progression. The bones underlying the ulcer site are often susceptible to contiguous osteomyelitis when bacteria devitalise the superficial cortex and extend the infection via the Haversian system into medullary bone and marrow (8). Here, arrest of the infection can be achieved with offloading, debridement and antibiotic therapy. However, hematogenous osteomyelitis may result when bacteria gain access into the bloodstream and seed into distal bones such as the vertebra (6). Spinal epidural abscesses are rare and classically present with back pain, fever and motor weakness. Positive blood and wound cultures, raised inflammatory markers together with a spinal MRI scan can confirm diagnosis. Cases of metastatic abscesses with infected diabetic foot ulceration as the source have been reported, but less commonly reported are epidural abscesses as distal metastatic sites (7). Spondylodiscitis has been shown to be difficult to diagnose and according to previous reports the delay in diagnosis may last up to 2–6 months (9). Since spondylodiscitis is uncommon, no randomised controlled trials of antibiotic therapy have been done and the consensus on disease management is still unclear.
At presentation, the foot ulcer was suggestive of soft tissue infection. The patient’s history of osteomyelitis together with backpain in the absence of trauma prompted spinal imaging. In this case, the most likely source of spinal infection was from the chronically infected diabetes foot ulcer, supported by blood culture findings and lack of alternative infection source. Prompt diagnosis and aggressive management are critical to the prognosis and final outcome. However, there is little data available on diabetic foot ulcerations as a primary site responsible for vertebral osteomyelitis. The interval between the patient’s initial contact with the diabetes foot service and development of signs of spondylodiscitis was 17 months. During this time, she had defaulted from the diabetes foot service, which prevented continuity of care, regular expert foot assessment and made her antibiotic concordance difficult to assess. The non-attendance together with poor glycaemic control and the significant comorbidities could have had a far-reaching impact on prompt diagnosis and outcome. However, diabetes foot service appointment non-attendances are common and represent missed opportunities to monitor wound healing and antibiotic therapy concordance (10). This raises the important question of how best to encourage concordance with outpatient management, and whether earlier inpatient management in poorly concordant patient would reduce morbidity and mortality in diabetic foot disease.
This case report highlights the role of diabetes foot disease as a primary distant focus for vertebral osteomyelitis, which can be a late and under-recognised complication of diabetic foot disease. It emphasises the importance of engaging with the multi-disciplinary diabetes foot service to improve short and long-term outcomes.
Declaration of interest
The authors declare that there is no conflict of interest that could be perceived as prejudicing the impartiality of the research reported.
Funding
This research did not receive any specific grant from any funding agency in the public, commercial or not-for-profit sector.
Patient consent
Informed consent was obtained from the patients’ NOK for the publication of this article.
Author contribution statement
A S and T T contributed to conception, interpretation, drafting, revision and final approval. S B, C N J and V B contributed to conception, interpretation and final approval of work. All authors agree to be accountable for all aspects of the work and contributed significantly. V B was the main physician responsible for the patient who granted permission for the case report.
References
- 1↑
Mavrogenis AF, Megaloikonomos PD, Igoumenou VG, Panagopoulos GN, Giannitsioti E, Papadopoulos A & Papagelopoulos PJ Spondylodiscitis revisited. EFORT Open Reviews 2017 2 447–461. (https://doi.org/10.1302/2058-5241.2.160062)
- 2↑
Jneid J, Lavigne JP, La Scola B & Cassir N The diabetic foot microbiota: a review. Human Microbiome Journal 2017 5 – 6 1–6. (https://doi.org/10.1016/j.humic.2017.09.002)
- 3↑
Euba G, Narváez JA, Nolla JM, Murillo O, Narváez J, Gómez-Vaquero C & Ariza J Long-term clinical and radiological magnetic resonance imaging outcome of abscess-associated spontaneous pyogenic vertebral osteomyelitis under conservative management. Seminars in Arthritis and Rheumatism 2008 38 28–40. (https://doi.org/10.1016/j.semarthrit.2007.08.007)
- 4↑
Robach BP, Niethammer TR, Paulus AC, Melcher C, Birkenmaier C, Jansson V & Wegener B Surgical treatment of patients with spondylodiscitis and neurological deficits caused by spinal epidural abscess (SEA) is a predictor of clinical outcome. Journal of Spinal Disorders and Techniques 2014 27 395–400. (https://doi.org/10.1097/BSD.0000000000000030)
- 5↑
Mantovani A, Trombetta M, Imbriaco C, Rigolon R, Mingolla L, Zamboni F, Dal Molin F, Cioccoloni D, Sanga V & Bruti M et al.Diabetic foot complicated by vertebral osteomyelitis and epidural abscess. Endocrinology, Diabetes and Metabolism Case Reports 2016 2016 150132. (https://doi.org/10.1530/EDM-15-0132)
- 6↑
Ceilley RI Foot ulceration and vertebral osteomyelitis with Corynebacterium haemolyticum. Archives of Dermatology 1977 113 646–647. (https://doi.org/10.1001/archderm.1977.01640050106018)
- 7↑
Nicolosi N & Pratt C Infectious spondylodiscitis, epidural phlegmon, and psoas abscess complicating diabetic foot infection: A case report. Journal of Foot and Ankle Surgery 2016 55 267–271. (https://doi.org/10.1053/j.jfas.2014.06.022)
- 8↑
Sadeghpour Heravi F, Zakrzewski M, Vickery K, G Armstrong D & Hu H Bacterial diversity of diabetic foot ulcers: current status and future prospectives. Journal of Clinical Medicine 2019 8 1935. (https://doi.org/10.3390/jcm8111935)
- 9↑
Lebre A, Velez J, Rabadão E, Oliveira J, da Cunha JS & Silvestre AM Infectious spondylodiscitis. Infectious Diseases in Clinical Practice 2013 1 223–228. (https://doi.org/10.1097/IPC.0000000000000118)
- 10↑
Bullen B & Young M Unmasking the impact of clinical non-attendance on neuropathic diabetes foot ulcer management. Diabetic Foot Journal 2019 22 28–33.