Applying negative pressure therapy to deep pressure ulcers covered by soft necrotic tissue

Abstract

Negative pressure therapy was applied to 39 deep pressure ulcers covered by soft necrotic tissue. All of the wounds were so deep that there were bones or ligaments just beneath the soft necrotic tissue. They had already received several types of conservative therapy including a necrotomy for periods ranging from 1 to 72 months. The wounds were minimally debrided and put in an adequately wet environment with negative pressure. This environment was established by the application of a suction‐dressing. All of these wounds showed clear wound beds within only 1 month. All of the wounds were successfully cured, either with or without a musculocutaneous flap. Negative pressure wound therapy is thus considered to be one of the choices for the management of non healing deep pressure ulcers covered by soft necrotic tissue.

INTRODUCTION

Pressure ulcers are frequently covered by soft necrotic tissue. Although debridement of the necrotic tissue is said to be necessary for the management of the wounds, sufficient surgical debridement is sometimes limited by the surrounding hard tissue including bones or ligaments. Even if sufficient debridement can be performed by an enzymatic debrider, it also inhibits healthy granulation which thus makes ulcers larger. As a result, daily wound cleansing and dressing changes should be continued for a long period of time in such cases.

Although a wound covered by necrotic tissue was conventionally regarded as contraindication of the application of negative pressure therapy (1), deep pressure ulcers covered by soft necrotic tissue were very successfully managed with minimal debridement followed by negative pressure therapy in a recent consecutive series of 39 wounds.

METHODS

The patients were limited to those who had a pressure ulcer bed which consisted of bone, ligaments or tendons covered by soft necrotic tissue. The destruction of the joint capsule and the existence of tunnelling were not excluded. Such patients were classified as grade IV ulcers according to the classification recommended by the National Pressure Ulcer Advisory Panel (2) (Figure 1A, B).

Figure 1.

(A) A deep pressure ulcer on the sacral portion. Soft necrotic tissue adhering to the bone and ligaments is not suited for the surgical debridement. (B) A deep pressure ulcer on the trochanter portion. The joint capsule is damaged and replaced by a large amount of necrotic tissue. Resection of the soft necrotic tissue is stopped when a little oozing haemorrhage is observed. The negative pressure therapy is initiated although some visible necrotic tissues still remain.

The study enrolled 32 patients who entered our hospital between August 2005 and July 2009 who showed 39 such wounds in all as mentioned above.

The medical ethics board of Kanoiwa Hospital approved the study. Informed consent was obtained from all patients before inclusion in this clinical study.

Minimum surgical debridement was performed until oozing haemorrhage was observed among the necrotic tissue. The total visible necrotic tissue was not rigorously resected to prevent a significant haemorrhage. The exposed bone as well as the thin soft necrotic tissue adhering to the bones or ligaments remained. Thereafter, a suction‐dressing was applied as the negative pressure wound therapy. The dressing contained a sponge, an evacuation tube and an adhesive drape. The sponge is made by polyurethane foam and widely marketed as a kitchen utensil (Figure 2A). The sterilised sponge was placed onto the wound defect which was covered by soft necrotic tissue. The evacuation tube was then embedded in the sponge. The surface of the wound containing the sponge was covered with an adhesive drape. The tube was connected to a suction bag developed for surgical drainage (SB bag^®, Sumitomo Bakelite Co., Japan or a HAMA^® drainage system, Hama Co., Japan; Figure 2B, C). The suction bag applied a suction pressure of 70–120 mmHg all day. The dressings were changed at 24–72 hour intervals according to the amount of necrotic tissue. The dressings were changed at short intervals especially in the early stages of treatment, because the cells of the sponge were likely to become clogged with necrotic tissue. In addition, some fragile necrotic tissues were removed during these dressing changes.

Figure 2.

(A) A sterilized polyurethane foam sponge made as a kitchen utensil was cut into an adequate size. (B) The suction‐dressing using an SB bag^®. (C) The suction‐dressing using the HAMA^® drainage system.

The wounds measuring over 20 cm² were closed with either a skin graft or a musculocutaneous flap when they appeared to be clear of necrotic tissue as far as the patient's condition permits. Small wounds measuring less than 20 cm² and large wounds in patients in poor condition were managed with the dressing regimen followed by the application of appropriate ointment.

Most of the patients discharged from the hospital after the completion of the wound management. The patients' safety and recurrence of the ulcer were determined by a telephone interview and the author checked directly by himself when it was considered to be necessary.

Histological observation

The sponge surface contacting the wound, which is shown in Figure 3B, was submitted for histological observation. The sponge surface was fixed in 20% formalin, paraffin‐embedded, sectioned vertically to the surface and stained by haematoxylin–eosin.

Figure 3.

(A) The same patient in Figure 1A. The necrotic tissue disappeared 3 weeks after the initiation of suction‐dressing, and the bone and ligaments were covered with the healthy granulation tissue. (B) The same patient in Figure 1B. A considerable amount of necrotic tissue disappeared 3 weeks after the initiation of suction‐dressing, and the joint capsule was covered with healthy granulation.

RESULTS

The 32 patients enrolled in this series had a total of 39 wounds (Table 1). The mean age was 82·4 (range, 71–95 years). They all were confined to bed due to stroke in 15 patients, respiratory disorders in 10, heart failure in 4 and dementia in 3. Two of the patients were associated with diabetes mellitus.

Table 1.

Profiles of our cases (32 patients with 39 ulcers)

	Patient age	Ulcer region	Duration spent before the initiation of NPWT (days)	Duration of NPWT (days)	Duration using ointments (days)	Days from the initiation of NPWT until the completion of the wound management	Basic disease	Follow‐up periods after the completion of the management (days)	Recurrence
NPWT alone: 16 patients with 19 ulcers
	92	Sacral	132	D *			Pneumonia
		Trochanter	186
	90	Sacral	58	D *			Congestive heart failure
		Heel	28
	94	Sacral	89	D *			Pneumonia
		Elbow	133
	88	Sacral	91	D *			Congestive heart failure
	71	Thoracic spinous	87	D *			Pneumonia
	84	Sacral	168	28	D †		Stroke
	86	Sacral	226	20	D †		Stroke
	78	Sacral	304	26	D †		Stroke
	80	Sacral	109	32	D †		Stroke
	86	Sacral	146	30	D †		Pneumonia
	90	Sacral	347	91	21	112	Stroke	1311	Erosion
	95	Sacral	84	66	18	84	Pneumonia	448 (lost by pneumonia)	None
	88	Heel	54	18	14	32	Stroke	1157	None
	77	Elbow	62	21	8	29	Congestive heart failure	667 (lost by pneumonia)	None
	82	Thoracic spinous	109	30	20	50	Pneumonia	239 (lost by pneumonia)	Erosion
	89	Heel	119	18	13	31	Dementia	218	None
Mean	85·6		133·3	34·5	15·7	56·3		673·3
±SD	6·6		83·5	23·0	4·9	34·3		466·3
NPWT followed by skin graft or musculocutaneous flap operation: 16 patients with 20 ulcers
	73	Trochanter	79	10	14	24	Stroke	1576	None
	79	Sacral	79	28	16	44	Stroke	1449	Erosion
	78	Iliac crest	258	28	42	70	Stroke, diabetes mellitus	1311	None
		Iliac crest	208	31	10	41		1124	None
	77	Sacral	129	14	14	28	Stroke	1066	Erosion
	79	Trochanter	61	16	10	26	Stroke, diabetes mellitus	1022	None
	80	Trochanter	54	17	12	29	Stroke	935	None
	82	Trochanter	99	20	12	32	Pneumonia	857	None
	84	Sacral	144	24	14	38	Congestive heart failure	360 (lost by stroke)	None
		Iliac crest	79	24	14	38		360 (lost by stroke)	None
	86	Sacral	2154	25	14	39	Stroke	673	None
	85	Trochanter	728	30	14	44	Pneumonia	557	None
	87	Sacral	1072	35	16	51	Stroke	498	None
	73	Sacral	611	36	25	61	Pneumonia	446	None
	82	Trochanter (right)	251	30	14	44	Pneumonia	404	None
		Trochanter (left)	247	28	14	42		386	None
	75	Sacral	48	12	90 ‡	102	Stroke	160	None
	71	Trochanter (right)	246	20	46 ‡	66	Dementia	78 (lost by pneumonia)	None
		Trochanter (left)	178	34	38 ‡	72		72 (lost by pneumonia)	None
	76	Sacral	88	26	10	36	Dementia	83	None
Mean	79·2		340·7	24·4	22·0	46·4		670·8
±SD	4·9		501·6	7·7	19·3	19·3		469·0
Summary of the 32 patients with 39 ulcers
	Age: 82·4 ± 6·6
	Duration spent before the initiation of NPWT (days): 239·6 ± 374·3
	Duration of NPWT (days): 28·0 ± 15·4 (five patients died before completion of NPWT.)
	Ulcer region: sacral 19, trochanter 10, heel 3, elbow 2, thoracic spinous 2, iliac crest 3
	Basic disease: stroke 15, pneumonia 10, congestive heart failure 4, dementia 3 (Diabetes mellitus was associated in two.)

NPWT, negative pressure wound therapy.

^*Died before completion of NPWT.

^†Discharged before completion of the ointment therapy.

^‡Post‐surgical additional application of the suction‐dressing to the dehiscenced wounds.

Moreover, 19 of the 39 pressure ulcers were in the sacral portion, 10 in the trochanter, 2 in the elbow, 3 in the foot, 3 at iliac crest and 2 were at the thoracic spinous prominence. The wounds had been managed with daily ordinary wound care including surgical debridement during periods ranging from 1 to 72 months (28–2154 days, average 240 ± 374 days) without favourable results.

The suction‐dressing was applied as negative pressure wound therapy with or without minimal surgical debridement. Five patients in poor condition with eight large wounds showed a deterioration of their original disease (pneumonia: three, heart failure: two) and died before they were able to achieve complete wound healing. The suction‐dressing was applied 28·0 ± 15·4 days in the remaining 27 patients. All of their 31 wounds showed clear wound beds even in the cases with exposed bone or a damaged joint cavity (Figure 3A, B). Although two cases required antibiotics for the inflammation around the wound which was already observed before the application of the suction‐dressing, there were no cases that showed a systemic fever due to inflammation around the wound after the application of the suction‐dressing.

Moreover, 16 patients with 20 wounds measuring over 20 cm² were treated by a skin graft or a musculocutaneous flap when the necrotic tissue was determined to have been sufficiently removed. These 16 patients with 20 wounds underwent 24·4 ± 7·7 days of negative pressure wound therapy followed by surgical management with a skin graft or a musculocutaneous flap. The total curative time was 46·4 ± 19·3 days. One patient with very large sacral wound (over 400 cm²) required 90 days and two large wounds at trochanters (over 80 cm² at each) in another patient required 38, and 46 days for each after the flap surgery because of the dehiscence of the surgical wounds. These wounds were managed with post‐surgical additional negative pressure wound therapy. Two wounds of these 20 wounds showed erosion at the same portion as the managed wounds but no cases showed a recurrence of the deep ulcers during the follow‐up of 671 ± 469 days (83–1576 days).

The other 11 patients with 11 wounds under 20 cm² received negative pressure wound therapy 34·5 ± 23·0 days until the soft necrotic tissue was replaced by healthy granuloma tissue and epithelialisation occurred at the wound boundary. Five of these patients were discharged before the completion of the wound healing because they showed good healing and needed no more negative pressure wound therapy. The remaining six patients with six wounds received further management with ointment for 15·7 ± 4·9 days. The total curative time in these six wounds was 56·3 ± 34·3 days. Erosion occurred on two of these six healed wounds during the follow‐up of 673 ± 466 days (218–1311 days).

Histological observation showed active granulation tissue with rich capillary beds in the spaces of the sponge trabeculae. The granuloma tissue was also associated with countless neutrophils and many macrophages (Figure 4).

Figure 4.

The sponge covering the wound shown in Figure 3B was submitted for the histological observation 2 days after exchange. A vertical section of the sponge surface (upper right) was observed. The sponge was dissolved during the staining and the wound tissue in the sponge mesh was left (haematoxylin–eosin stain). (B) High‐power view of the same section as in (A). Note the marked capillary proliferation and the countless neutrophils and macrophages (haematoxylin–eosin stain).

DISCUSSION

A suction‐dressing device was developed for negative pressure wound therapy by Argenta and Morykwas in the late 1990s and applied for chronic, non healing wounds without infection.(3) The suction‐dressing device was developed as a commercially available VAC^® (Vacuum‐Assisted Closure, KCI, USA) 3, 4, which applies a vacuum pressure of approximately 125 mmHg, thereby creating a subatomospheric environment. The VAC^® device provides several advantages. Perturbation of the extracellular and intracellular balances produces cell extension and spreading which activates cell proliferation and matrix formation of the wound bed tissue. 5, 6, 7, 8, 9 In addition, the bacterial cleaning power derived from the aspiration of the contaminated fluid stimulates the wound‐healing process. (4) Furthermore, the removal of chronic wound fluid, which inhibits cell proliferation (10) allows both cellular proliferation and tissue granulation. Finally, the blood flow to the wound also increases in such a subatomospheric environment. (4) Although the commercially developed devices, including the VAC^®, are not available in Japan, their advantages can be safely provided by a simple suction‐dressing device (11) and is useful to clean the necrotic tissue which is assumed to be generated by bacterial infection and local ischaemia.

The wounds covered with soft necrotic tissue displayed a clear wound bed in a shorter time than had originally been expected, based on their past long‐term treatment with daily wound care and dressing changes. The mean duration of management before the suction‐dressing application was 240 ± 375 days. Several mechanisms of clearance of the necrotic tissue during the suction‐dressing can be considered. They include repeated surgical resection of fragile necrotic tissue at the time of dressing changes, suction away after autolysis and removal of small necrotic tissue clogged in the sponge meshes. Another mechanism may include the living reaction of the granuloma in the inter‐trabeculae spaces of the sponge. The inflammatory cells along with macrophages are supported by the rich capillaries in the spaces, which serve to eradicate necrotic tissue and suppress bacterial propagation.

Six patients with six wounds received the suction‐dressing and the further management with ointment necessitated a total curative time of 56 ± 34 days. The other 16 patients with 20 wounds required a total curative time of 46 ± 19 days, including the suction‐dressing and the following skin graft or musculocutaneous flap operation. The total curative time was within a maximum of 4 months, even including the wounds that had previously received 72 months of daily management.

Although there has been one reported case of toxic shock syndrome during the treatment of an infected wound by such suction‐dressing (12), the current consecutive cases did not show any adverse effect which was assumed to be originated from the suction‐dressing.

Fortunately, there was no recurrence of deep ulcers during the observation periods even in cases with exposed bone or a damaged joint cavity. Invisible osteomyelitis should be considered if the granuloma does not grow to cover the bone within a few weeks. As a result, the strategy of performing minimal surgical debridement could be abandoned and a rigorous resection, including the bone, could thus be performed if necessary.

Further investigations should be conducted including a randomised prospective study in order to determine how fast a chronic non healing deep pressure ulcer covered by soft necrotic tissue heals with treatment with negative pressure wound therapy in comparison to ordinary methods.

ACKNOWLEDGMENT

My special thanks to Akira Kawaoi MD, Division of Pathology and Laboratory Medicine, for the helpful comments and suggestions.