References

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 https://doi.org/10.1001/jamadermatol.2019.0884

British Lymphology SocietyLS consensus doc 2022. Guidelines on the management of cellulitis in lymphoedema. 2022. https://www.thebls.com/documents-library/guidelines-on-the-management-of-cellulitis-in-lymphoedema (accessed 6 March 2023)

Kilburn SA, Featherstone P, Higgins B, Brindle R. Interventions for cellulitis and erysipelas. Cochrane Database Syst Rev. 2010; 2010:(6) https://doi.org/10.1002/14651858.cd004299.pub2

Sawai J, Hasegawa T, Kamimura T Growth phase-dependent effect of clindamycin on production of exoproteins by Streptococcus pyogenes. Antimicrob Agents Chemother. 2007; 51:(2)461-467 https://doi.org/10.1128%2FAAC.00539-06

Vignes S, Poizeau F, Dupuy A. Cellulitis risk factors for patients with primary or secondary lymphedema. J Vasc Surg Venous Lymphat Disord. 2022; 10:(1)179-185.e1 https://doi.org/10.1016/j.jvsv.2021.04.009

Wounds UK. Patients presenting with ‘red legs’: differential diagnosis and the role of compression. 2022. https://www.woundsme.com/uploads/resources/97f654a22b9afbac3bc6c9dac73e49d3.pdf (accessed 6 March 2023)

The management of acute cellulitis in individuals with lymphoedema: a case study

01 April 2023

Cellulitis is an acute spreading inflammation of the skin and subcutaneous tissues, characterised by pain, warmth, swelling and erythema (British Lymphology Society (BLS), 2022).

In a study focusing on 1846 patients attending a specialist lymphoedema centre, 37.6% reported having experienced at least one episode of cellulitis, while 23.3% noted having had recurrent cellulitis (Vignes, 2022).

In individuals with lymphoedema, cellulitis can cause severe systemic upset, with high fever, rigors and even sepsis, while others may only experience mild symptoms, with minimal or no fever. The volume of oedema may increase and inflammatory markers (C-reactive protein (CRP), erythrocyte sedimentation rate (ESR)) may be raised.

A Cochrane review and subsequent partial update concluded that, in general, based on existing evidence, it was not possible to define the best treatment for cellulitis (Kilburn et al, 2010; Brindle et al, 2019). Furthermore, the appropriate treatment of cellulitis in lymphoedema may differ from cellulitis in other clinical situations. The BLS consensus document (2022) for the management of cellulitis in individuals with lymphoedema makes recommendations about the use of antibiotics and advises when admission to hospital is indicated. Prompt treatment is essential to reduce the risk of worsening symptoms and the development of life-threatening conditions, such as sepsis. Furthermore, this could potentially stop any further damage to the lymphatics of the affected part, which may be predispose to repeated attacks. The consensus document can be accessed at: https://www.thebls.com/documents-library/guidelines-on-the-management-of-cellulitis-in-lymphoedema

Case study

A 36-year-old female (permission granted by patient to use them for this case study) with long-standing right leg lymphoedema (possibly secondary to joint injections performed in hospital, when she was 7 years old). The right leg swelling extended from toe to below the knee; there were no associated skin changes and the shape was normal. The swelling did not reduce overnight and there was an increase in oedema during her first pregnancy aged 34 years. The patient also has juvenille idiopathic arthritis and is on biologics (etanercept 50 mcg subcutaneous, once weekly), and she works part-time as a general practitioner (GP). She is married and has an 18-month-old daughter.

The patient's oedema was monitored overtime using excess limb volume (ELV). Overall, the ELV was 304 mls (2%); however, the distal ELV was 700 mls (13%) and proximal ELV was 397 mls (4%) (Table 1). Tissue dielectric constant (TDC) moisture meter measurements were also recorded; these were all within normal limits, except for the reading measured at the dorsum of the right foot (52%).


Table 1: Results of patient's oedema, which was monitored overtime using excess limb volume (ELV)
Measurements taken on two occassions (cm around hand/foot) 11/05/2021 20/07/2021
Right Left Right Left
26.5 23 26.6 25
Distal        
1 35 33.3 31.0 27
2 31 26.2 35.2 29.3
3 34.3 29.5 39 32
4 39.2 32.5 42.4 37.5
5 42.2 38.6 45.9 43
6 46.8 44 47.5 46
7 48 44.4 48.2 45.5
8 47.0 45.3 47 45.5
9 46 45.2 46 43.8
10 44 43 44.9 48
11        
12        
Distal total 5550 4795 5903 5203
Proximal        
1 43.5 48 52 51.3
2 49 53 55.5 56.4
3 54 57.8 59 61
4 59.5 63.9 64.6 66.5
5 64 66.3 68.6 70.3
6 67.3 70 73 72
7 72 75.5 76.5 78.6
8 73.2 75.7 77 79
9        
10        
11        
12        
Proximal total 9523 10588 11222 11619
Total mls 15073 15383 17125 16822
Excess mls 310 304
Total Excess limb volume % 2% 2%
Distal mls 755 700
Distal % 16% 13%
Proximal mls 1065 397
Proximal % 11% 4%
PD Ratio 0.5 0.3
ELV difference from last data   2.1%
Right limb difference from last data   -13.6%
Left limb difference from last data   -9.4%

Her previous lymphoedema treatment included decongestive lymphatic therapy on three occasions, along with skincare and exercise. She currently wears compression daily in the form of made-to-measure, flat-knit compression hosiery (class 3 forte; Elvarex), ReadyWrap toe cap and night time compression (Comfiwave).

Since her early 20s, the patient has experienced right lower limb cellulitis on a recurrent basis. In 2007, she was admitted for IV treatment and also commenced on long-term prophylaxis (18 months) of phenoxymethylpenicillin, which was effective for several years. Further episodes of infection were treated with flucloxacillin; however, she struggled to complete the course due to the onset of bloody diarrhoea around day 3-4, therby, taking clarithromycin as an alternative. Nonetheless, with calrithromycin, she would often need repeat and multiple courses to completely clear the cellulitis. Her last episode of cellulitis was 5 years ago and following a course of flucloxacillin and two courses of clarithromycin, her GP decided to administer a 10-day course of co-amoxiclav, which treated the infection and led to a 4-year absence of recurrence.

In October 2022, she sustained a tiny cut to her 4th toe on the previously affected limb, which was promptly cleaned and dressed. Yet, 2 days later, she started to feel unwell, developing a fever and sweats. Initially, she thought that she was getting a cold-her daughter had recently been unwell-but after a few hours she experienced rigors, nausea and pain in her right leg around the knee and thigh (no increase in oedema). This presentation is typical for her cellulitis and the rash then appeared approximately 8-10 hours after the fever started. As soon as the development of the tiny red blanching macules appeared on her foot, she commenced her rescue antibiotics from home (which were out of date but rang 111 for an emergency replacement—sadly, this further delayed antibiotic administration by 14 hours). Despite this, the rash rapidly progressed and the entire right foot and lower leg up to the mid-shin became erythematous, painful, warm and more swollen with a few small satellite patches of erythema beyond the main affected area. The patient also began to vomit, but there was no diarrhoea; she felt thirsty and had noted less urine output. After 24 hours of antibiotics and no improvement, the patient performed her own observations at home. Her blood pressure was 91/61, heart rate (HR) was 132, temperature (temp) 39.6°c and felt very dizzy and pre-syncopal. The decision was made to attend A&E to rule out sepsis.

She presented in A&E, with a temp of 38.5°c, tachycardic HR of 137, and RR interval of 28. Bloods and cultures were taken and her CRP was 371 and estimated glomerular filtration rate (eGRF) was 46 (reduced from >90) (Figure 1). Sepsis was suspected and severe acute kidney injury (AKI) was diagnosed. Prior to the blood test results coming back, the patient was prescribed a stat dose of co-amoxiclav in A&E and also a stat bag of IV hartmans and paracetamol, which was effective at stabilising her HR and temp. As previously stated, the patient was intolerant of flucloxacillin, and had developed bloody diarrhoea in more than one occasion after being prescribed it. Once reviewed by the medical team with her blood test results at hand, the consultant initiated a plan of IV vancomycin (local guidelines for severe cellulitis but also due to abnormal liver function test, secondary to infection) and a further 3L of IV fluids. She was also referred to the infectious diseases consultant for review, who advised the addition of clindamycin 450 mg qds for 2 days (patient was pre-warned that only 48 hours' medication would be given, as it is poorly tolerated) and clotrimazole 1% cream for suspected athlete's foot (Table 2).

Figure 1. Patient leg when she first presented at A&E.

Table 2. Medications prescribed to patient after review by consultant
Name Route Dose Frequency Duration Prescriber notes
Chlorphenamine Oral 4 mgs when required 5 Urticarial rash
Paracetamol Oral 1 g when required 14 Fever
Dalteparin Subcutaneously 5000 units daily 7 Immobility due to cellulitis
Clotrimazole 1% cream Topical 1 twice daily 14 Suspected athlete's foot

Due to AKI, the patient's immunocompromised status and dehydration, she was admitted to the short stay unit and subsequently, the infectious diseases ward. The administration of vancomycin and clindamycin led to a reduction in erythema and tenderness and the patient's clinical condition improved significantly, including the complete resolution of the severe AKI. She developed intense full body sweats approximately 1 hour following each administration of clindamycin and then had a skin reaction on the 7th dose, where she developed a widespread urticarial rash. At this point, the last dose was omitted. After 2 days, the patient was medically fit for discharge, with a CRP of 240, blood cultures negative to-date (taken >24 hours after commencing oral antibiotics) and COVID-negative. The plan was to discharge on IV ceftriaxone with outpatient parenteral antimicrobial therapy (OPAT), but the patient was unable to be cannulated (17 attempts were made). The infectious diseases consultant prescribed Dalvance®dalbavancin 1.5 mg subcutaneously via a butterfly needle over 90 minutes.

Dalbavancin for injection is indicated for the treatment of adult patients with acute bacterial skin and skin structure infections caused by gram-positive microorganisms: Staphylococcus aureus, Streptococcus and Enterococcus faecalis. The recommended dosage in patients with normal renal function is 1500 mg, administered either as a single dose, or 1000 mg followed by 500 mg one week later, usually administered over 30 minutes by intravenous infusion.

The patient was pre-warned that if the dalbavancin failed to manage her symptoms (or they were unable to administer the full dose via butterfly needle) she would then be prescribed oral linezolid. Follow up was planned for repeat bloods in 4 days and clinical review 1 day later.

Once home, the patient managed the significant increase in swelling with a below knee Comfiwave garment worn for 24 hours. The swelling started to slowly resolve in the calf but foot remained very swollen, especially the toes. The redness was markedly reduced very quickly. Sequential pneumatic compression was tolerated but as the leg remained tender, compression hosiery could only initially be worn for an hour at a time. Due to extensive foot swelling, the skin began to extensively peel and was cracked and dry. Flexitol heel balm was employed by the patient with good effect, along with regular washing of the feet and further emollients.

Follow up blood tests showed CRP of 18.4. Infectious diseases consultant's review concluded prophylaxis in the event of a further attack and amoxicillin 1 g tds as a standby rescue medication. Day 8, the patient was able to return to wearing compression hosiery daily, but only 1 pair of shoes were well-fitting due to the increase in foot oedema. Figure 2 shows the patient's legs at Day 10.

Figure 2. Patient's leg 10 days post-cellulitis.

Discussion

The Wounds UK Best Practice Statement (2022) suggests that compression therapy should be considered in patients with cellulitis as appropriate to their underlying vascular status, with consideration given to the patient's pain levels and whether they can tolerate compression. If the patient is already established in compression therapy and develops cellulitis, compression therapy should not be stopped and should instead, be continued at the highest level that the patient can tolerate.

Despite this growing trend to encourage patients to wear compression hosiery daily and continue to do so during an acute cellulitis event, in this case, the patient did not feel well enough to remember to take her compression hosiery into hospital. She did not feel it was a priority while she was acutely unwell. The total episode of time without hosiery was 4 days and she was largely on bed rest for this time. Her current prescription of made-to-measure garments would not have fitted due to the volume increase in her swelling. Furthermore, to have measured and ordered new garments to accommodate the increase in oedema would have been ineffective as the garments would not have been received on time (average time for provision of made-to-measure garments is 5–10 days).

The BLS consensus document recognises the difficulties encountered by individuals with cellulitis and lymphoedema (BLS, 2022). If wearing the usual compression garment causes pain, then it should be removed, but replaced as soon as the affected area is comfortable enough to tolerate it. This should reduce the risk of worsening of the swelling if the garment is left off for a prolonged period (e.g., one week). The fit of the compression garment may need to be checked as the area may become more swollen after an episode of cellulitis (BLS, 2022).

The patient described the Comfiwave garment in the first few days of recovery as:

‘absolutely perfect, so gentle and soft, not tight but giving exactly the right level of support and comfort.’

She described her leg as feeling snug and protected, and would recommend the garment for all individuals experiencing recurrent cellulitis. Due to the rapid onset and often, quick resolution of swelling in patients with lymphoedema and cellulitis, a soft and accommodating garment may be more effective during the acute phase. She also reported that, following the application of her usual compression garments, the foot swelling reduced more rapidly than usual.

According to the patient, the redness resolved more rapidly than she had noted in previous episodes of acute cellulitis. It is suggested that the proliferation of bacteria in streptococcal infection can lead to the release of exotoxins, which can cause tissue damage and also damage the endothelium, leading to fluid leakage, tissue swelling and erythema. Although more research is required, clindamycin is thought to have efficacy against the production of bacterial exoproteins (Sawai, 2007). The patient was prescribed clindamycin that is thought to have had an effect on the exotoxins, which, in this case, led to the development of fever and rigors before the redness/rash appeared. Amoxycillin is considered to not be resistant to streptococcal infection but clarithromycin can be; hence, why the patient previously required multiple courses of antibiotics.

It is necessary to consider that the faster resolution of redness, with multiple patches appearing where the redness had faded very quickly, may indeed be related to the clindamycin. However, the patient has previously managed on oral antibiotics and the rapid improvement may be related to the intravenous vancomycin.

Given the significant and rapid improvement to treatment, the patient would like to have access to intravenous therapy for any future episodes of cellulitis and she has been given open access for review by the OPAT team.

Conclusion

The case study highlights that the onset and presentation of cellulitis can be very different, not only amongst individuals, but within the same individual, and management should always be individualised. The patient, in this case, would not be recommended to have prophylactic antibiotics as the consensus document reserves this for those experiencing two or more episodes of cellulitis in a 12-month period. It is clear that although hospital management of cellulitis is according to local policy, a step down to oral antibiotic therapy, treatment should be extended for 14, rather than 7, days.