Changes in sternal perfusion following internal mammary artery bypass surgery

Abstract

BACKGROUND:

Coronary bypass surgery using the internal mammary artery (IMA) is among the most commonly performed procedures in treatment of advanced coronary vessel disease. Further, bilateral harvesting of the IMA is associated with increased rates of sternal wound infections.

OBJECTIVE:

This study aimed to explore changes in sternal perfusion, following left internal mammary artery (LIMA) harvesting.

METHODS:

60 patients were divided equally into intervention- and control cohorts and underwent assessment of oxygen saturation (sO2) and relative blood flow (RBF) by laser doppler flowmetry remission spectroscopy preoperatively, 24 h and 72 h postoperatively.

RESULTS:

24 h postoperatively a significant decrease of sO2 and RBF could be detected in LIMA harvest patients. After 72 h this difference disappeared. The side comparison within the groups 24 h postoperatively revealed a reduction of sO2 on the LIMA side within both groups. Regarding the 24 h RBF in side comparison, no significant differences could be detected within the groups. The 72 h side comparison showed an increased RBF for the non-LIMA side within the control group.

CONCLUSIONS:

The use of the LIMA may lead to significant decreases in local tissue oxygen saturation and RBF. This reduction is mostly distinct within the first 3 days postoperatively and may influence sternal wound infections.

Keywords

Sternal wound infection LIMA sternal infection sternum perfusion tissue oxygenation coronary bypass surgery

1 Introduction

In the past decades the internal mammary artery (IMA) has been well recognized as vascular graft source with superior patency rates in comparison to vessels from other donor sites in general and venous grafts in particular. With regard to this specific donor source, vascular bypass surgery using the IMA emerged to one of the most commonly performed procedures in treatment of advanced coronary vessel disease [1 –3]. Nevertheless, the harvest of the IMA, in particular performed bilaterally, is associated with increased rates of sternal wound infections, a severe complication following cardiac surgery [4, 5]. Particularly deep sternal infections cause high morbidity, increased mortality, prolonged hospital stay and considerably boost treatment expenses [5 , 17]. Treatment of this life-threatening condition is an interdisciplinary challenge, making advanced microsurgical procedures necessary, especially in extensive disease [19 –21]. According to the literature the incidence of sternal wound complications following median sternotomy during cardiac surgery is as high as 8% with a mortality rate ranging between 14 and 50% [4 , 7]. Besides bilateral IMA harvesting as the major risk factors for sternal infection, there are other known predictors, such as obesity, diabetes, smoking, peripheral vascular disease, chronic obstructive pulmonary disease and advanced age [5 –7]. The significant decrease of sternal blood circulation with failure of local microcirculation in particular, is believed to cause sternal vulnerability to infection [4, 16]. Despite this, as long as techniques like e.g. stem cell therapy for treatment of ischemic heart disease still remain experimental approaches, revascularization with autologous vascular bypass grafts remain the gold standard therapy [22]. Nevertheless, the role of perfusion and the presence of oxygen in wound healing have not been completely understood so far. However, numerous experimental and clinical trials have shown impaired wound healing and increased infection rates associated with hypoxic conditions [8–10 , 24] or in reverse, a better healing case of well perfused tissue [23]. This study aimed at exploring the changes in sternal perfusion, following harvest of the left internal mammary artery (LIMA) as coronary bypass graft using a combined white light photospectroscopy and laser Doppler device as a non-invasive technique, permitting an exact pre- and postoperative measurement of microcirculation and tissue oxygensaturation.

2 Material and methods

2.1 Patient cohort

Upon written informed consent a total of 60 patients were enrolled in this study. Study participants were allocated to either intervention- or control group (Table 1). Patients in the intervention cohort received coronary artery bypass grafting using the left internal mammary artery (LIMA). The control cohort consisted of patients undergoing cardiac surgery performed through median sternotomy without IMA harvesting (Table 1). Exclusion criteria in this study were denial of written patient consent, previous sternotomy, thoracic operations, procedures affecting the IMAs as well as emergency- and revision surgeries. Demographics and underlying diagnoses of both groups can be depicted in Table 1. For the most part the cohorts did not differ significantly in terms of demographics, secondary diagnosis and risk factors. Except the diagnosis of hyperlipidemia was significantly more prevalent within the intervention group (p = 0.00).

Table 1
Patient characteristics intervention-vs. control group in terms of absolute numbers (n) and p-values

Patient characteristics

Intervention Control P

Cohort size n = 30 n = 30

Sex (male/female) 23/7 17/13 0.10

Mean age (Years) 66 62.4 0.21

Type of surgery Coronary LIMA Median sternotomy without IMA

bypass (n = 30) harvesting

Aortic value reconstruction (n = 20)

Mitral value reconstruction (n = 3)

Tricuspid value reconstruction (n = 1)

Ascending aorts reconstrucion (n = 5)

Atrial myxoma removal (n = 1)

Diabetes mellitus n = 9 n = 9 1.00

Hyperlipidemia n = 18 n = 5 0.00

Hypertension n = 23 n = 18 0.60

BMI >30 n = 10 n = 9 0.79

Alkohol n = 16 n = 14 0.61

Nicotine n = 6 n = 7 0.38

Patient characteristics
Cohort size	n = 30	n = 30
Sex (male/female)	23/7	17/13	0.10
Mean age (Years)	66	62.4	0.21
Type of surgery	Coronary LIMA	Median sternotomy without IMA
	bypass (n = 30)	harvesting
		Aortic value reconstruction (n = 20)
		Mitral value reconstruction (n = 3)
		Tricuspid value reconstruction (n = 1)
		Ascending aorts reconstrucion (n = 5)
		Atrial myxoma removal (n = 1)
Diabetes mellitus	n = 9	n = 9	1.00
Hyperlipidemia	n = 18	n = 5	0.00
Hypertension	n = 23	n = 18	0.60
BMI >30	n = 10	n = 9	0.79
Alkohol	n = 16	n = 14	0.61
Nicotine	n = 6	n = 7	0.38

2.2 Data acquisition

The measuring apparatus (oxygen-to-see (O2C)) is a flowmetry and remission spectroscopy system (LEA Medizintechnik, Giessen) [11,18, 11,18]. The measuring unit, which is connected to an optical fiber probe (LF2 flat probe, LEA Medizintechnik, Giessen), was placed on the skin surface of the patient‘s sternum at four different measuring points (M1-M4), 1.5 cm lateral of where the conventional fashioned midline sternotomy was performed. The height localization was determined between the superior and middle third, as well as between the middle and the inferior third of the sternum length. Therefore two measuring points (M1, M2) were located on the right sternal side, whereas two measuring points (M3, M4) were situated on the left side of the sternum, where the LIMA harvesting was performed within the bypass group (Fig. 1).

Fig.1

Measuring points.

The measurements were carried out preoperatively, 24 h and 72 h postoperatively in each patient. The probe was encased by a sterile cover (Ultracover, Cardinal Health Germany GmbH) and placed on the skin whereby the O2C device recorded tissue oxygen saturation (sO₂), relative blood flow, blood flow velocity and relative amount of haemoglobin (rHb) in the measured tissue. Except the tissue oxygen saturation shown in %, all parameters are expressed in arbitrary units (AU).

2.3 Statistical analysis

Data are expressed as median values (M) with interquartile range (IQR). Normal distribution was verified by viewing data within a Q-Q plot. Unpaired student‘s t-Test, was conducted to compare the two groups (control vs. intervention). Paired student‘s t-Test was used to investigate differences between the right and the left sternum within groups. Statistical significance was defined as value of p < 0.05, very significant as p < 0.01 and highly significant as p < 0.001. Analysis was carried out using Superior Performing Software System (SPSS), Version 22 for Windows (SPSS, Inc.,Chicago, III).

3 Results

3.1 Tissue oxygen saturation bypass- vs. control group

Tissue oxygen saturation (sO₂) at corresponding measuring points (e.g. M1 bypass- vs. M1 control group etc.) was compared between bypass and control group by unpaired student‘s t-Test. Equivalent preconditions were confirmed by preoperative measurements that suggested insignificant differences in between study groups (data not shown).

Twenty-four hours postoperatively a significant decrease of sO₂ could be detected for the bypass group in M4 when compared to controls (Fig. 2, Table 2).

At 72 hours this difference in sO₂ for M4 disappeared (data not shown).

3.2 Relative blood flow bypass- vs. control group

Similar data analysis was performed for relative blood flow, whereas equal preconditions could be confirmed for this parameter by the lack of preoperative statistical significant differences between the groups (data not shown).

In contrast, 24 hours postoperatively the measurements of relative blood flow revealed a significant decrease for M1, M3 as well as for M4 in the bypass group (Fig. 3, Table 3). Comparable to the change in sO2 this difference dissolved at 72 hours (data not shown).

3.3 Tissue oxygen saturation right vs. left sternum within groups

To investigate possible differences between the right and the left part of the sternum within study groups, M1 vs. M3 and M2 vs. M4 were compared by paired student’s t-Test.

Symmetrical oxygen saturation between right and left sternum was verified for both groups preoperatively (data not shown).

At 24 hours postoperatively this comparison within the bypass group revealed a significant decrease for M3 vs. M1 and a highly significant reduction of sO₂ for M4 vs. M2 (Fig. 4, Table 4).

Also a very significant decrease of M4 vs. M2 could be detected for the control group at 24 hours. Whereas this decrease was still significant at 72 hours within the control group, differences within the bypass group decreased to nearly significant (p = 0.052) for M4 vs. M2 (data not shown).

3.4 Relative blood flow right vs. left part of the sternum within groups

Similar data analysis was performed for relative blood flow and equal preconditions could be confirmed for this parameter by the lack of preoperative statistical significant differences within groups (data not shown).

At 24 hours after surgery no significant differences could be detected within the two groups. However, the bypass group showed a non-significant decrease for M4 vs. M2 (Fig. 5, Table 5).

At 72 hours postoperatively the bypass group showed a non-significant decrease of blood flow in M3 against M1 as well as M4 in comparison to M2 (data not shown).

Within the control group a very significant increase for M1 vs. M3 could be detected at 72 hours postoperatively, in contrast no significant difference was seen for M4 vs. M2 in this group (data not shown).

4 Discussion

Twenty-four hours postoperatively a significant decrease of sO₂ could be detected for the bypass group on the LIMA side when compared to controls. After 72 hours this difference in sO₂ disappeared. The 24 hours postoperative relative blood flow measurement revealed a significant decrease for the bypass group, predominantly on the left internal mammaria side compared to controls. Comparable to the change in sO₂, this difference of relative blood flow dissolved at 72 hours.

The side comparison within the groups 24 hours postoperatively revealed a reduction of sO₂ on the LIMA side within both groups, however this reduction was more distinct within the bypass group. Regarding the relative blood-flow in side comparison after 24 hours, no significant differences could be detected within the groups. The 72 hours side comparison showed an increased blood flow for the non-LIMA side within the control group.

Doing revascularization procedures the internal mammary artery (IMA) presented superior long-term patency rates in comparison to other donor vessels sources, and has therefore emerged to one of the most commonly used vessels in treatment of advanced coronary vessel disease [1 –3]. However, in particular the bilateral IMA harvesting is associated with increased rates of sternal wound infections compared to unilateral mammary artery grafts and in this manner with high patient morbidity, increased mortality, prolonged hospital stay and raised treatment expenses [5 , 17]. Due to the increased risk of diminished blood supply - related to sternal wound complications upon bilateral IMA harvest, this procedure is discussed controversially among cardiac surgeons and is generally not performed at this university hospital. However, until today it still remained unclear whether unilateral harvest of the LIMA may already result in changes of sternal blood perfusion, which may be prone to an increase of sternal wound complication rates, since the vulnerability to sternal wound infection has been shown to be related to a significant post-operative decrease of local blood supply [4, 16]. A more precise understanding of sternal microcirculation following median sternotomy and LIMA harvesting could therefore be helpful to reduce sternal wound healing complications including life threatening sternal osteomyelitis.

The measuring apparatus oxygen-to-see (O2C), a laser Doppler flowmetry and remission spectroscopy system, has delivered precise information about tissue oxygen saturation (sO₂), relative blood flow, blood flow velocity and relative amount of haemoglobin (rHb) in the measured tissue. In this study the focus was set on the parameters of oxygen saturation and relative blood flow in order to investigate changes in tissue microcirculation.

The 24 hours postoperative intervention-/control group comparison, with a significant decrease of sO₂ and relative blood flow on the left internal mammary side in the bypass group, demonstrates the expected significant IMA effect on sternal blood supply. After 72 hours this significant difference for sO₂ and relative blood flow between the two groups dissolved. This equalization might be based on tissue’s capacity to improve its microcirculation in times of ischemic conditions, commonly described as delay phenomenon. The delay phenomenon, also designated as vascular delay or ischemic preconditioning, describes the finding that tissues transposed into a partially ischemic state, will increase their perfusion by subsequently opening of choke vessels [12 , 15]. The increase of oxygen supply and relative blood flow noted at day three might have contributed to an appropriate wound healing, whereas the latter could have been impaired in conditions of disturbedmicrocirculation.

The side comparison within the groups 24 h postoperative showed a highly significant decrease of sO₂ on the LIMA side within bypass patients. This result substantiates the importance of the left internal mammary artery for sternal oxygen supply and might be among the predictors for sternal wound complications. Even if less distinct compared to the LIMA patients, the control group revealed a deficit of sO₂ on the left sternal side as well. This asymmetrical distribution within the control group might be explained by changes in flow conditions within the cardial outflow tract following valve or ascending aortic reconstructions.

The comparison in relative blood-flow between both sides within the bypass and the control group revealed no significant differences between the right and left sternum after 24 hours. However, the bypass group showed with exception of M2 significantly lower values for relative blood flow compared to the controls. This could imply a potential influence of the LIMA harvesting not only on the left, but also on the right sternal side. After 72 hours the side comparison showed a highly significant increase of relative blood flow in M2 within the control-group. This increase could be explained by the previously mentioned changes in cardial outflow conditions following valve replacements and aortic reconstructions respectively.

5 Conclusion

These findings indicate that the use of the left internal mammary artery as a bypass surgery graft may lead to significant decreases in tissue oxygen saturation and blood flow. This reduction is mostly distinct within the first three days after bypass surgery and may well explain the increased occurrence of sternal wound infections especially following bilateral IMA harvesting, as performed and described elsewhere. Previous studies, in which sternal microcirculation was investigated, had already shown a significant LIMA effect on sternal perfusion. In this study the measurements were carried out on the left-sided pre- and retrosternal bone during LIMA bypass surgery and revealed significant decreases of sternal oxygen saturation and blood flow after LIMA harvesting [14].

Further studies are needed to elucidate the mechanisms and dynamics of sternal microcirculation following the LIMA extraction and its utilization as vascular bypass graft.

The following limitations of this study need to be discussed critically: Since the measuring probe was placed on the skin surface, the data recorded reflect changes in blood flow and sO₂ within the skin and subcutaneous tissue above the sternum. In this respect, information about sternal bone perfusion parameters is based on the assumption, that suprasternal skin and sternum are subjected to similar vascular supply. It further needs to be noted that patients undergoing coronary artery bypass surgery tend to have poorer vascular status in general compared to patients within the control group. Since this circumstance might lead to a reduced comparability between the bypass- and the control patients, data were also analyzed in side comparison within the two groups.

Footnotes

Acknowledgments

The present work was performed as a part of Konstantin Klein’s doctoral thesis and was carried out in fulfillment of the requirements for obtaining the degree “Dr. med”. This degree is awarded by the Friedrich-Alexander-Universität Erlangen-Nürnberg (FAU).

References

Günday

, Coşkun

, Çiftçi

Ö.

, Özülkü

, Tekindal

M.A.

, Güven

, et al., Radial artery, saphenous vein versus left internal thoracic artery in recurrent ischemic symptoms after coronary artery bypass graft surgery, Arch Iran Med 17(8) (2014), 551–555.

Khot

U.N.

, Friedman

D.T.

, Pettersson

, Smedira

N.G.

, Li

and Ellis

S.G.

, Radial artery bypass grafts have an increased occurrence of angiographically severe stenosis and occlusion compared with left internal mammary arteries and saphenous vein grafts, Circulation 109(17) (2004), 2086–2091.

Shah

P.J.

, Bui

, Blackmore

, Gordon

, Hare

D.L.

, Fuller

, et al., Has the in situ right internal thoracic artery been overlooked? An angiographic study of the radial artery, internal thoracic arteries and saphenous vein graft patencies in symptomatic patients, Eur J Cardio-Thorac Surg Off J Eur Assoc Cardio-Thorac Surg 27(5) (2005), 870–875.

Hauser

, Steinau

H.U.

, Ring

, Lehnhardt

and Tilkorn

D.J.

, Sternal osteomyelitis. Etiology, diagnostics and operative therapy concepts, Chir Z Für Alle Geb Oper Medizen 85(4) (2014), 357–365; quiz 366–367.

Ridderstolpe

, Gill

, Granfeldt

, Ahlfeldt

and Rutberg

, Superficial and deep sternal wound complications: Incidence, risk factors and mortality, Eur J Cardio-Thorac Surg Off J Eur Assoc Cardio-Thorac Surg 20(6) (2001), 1168–1175.

Losanoff

J.E.

, Richman

B.W.

and Jones

J.W.

, Disruption and infection of median sternotomy: A comprehensive review, Eur J Cardio-Thorac Surg Off J Eur Assoc Cardio-Thorac Surg 21(5) (2002), 831–839.

Singh

, Anderson

and Harper

J.G.

, Overview and management of sternal wound infection, Semin Plast Surg 25(1) (2011), 25–33.

Schreml

, Szeimies

R.M.

, Prantl

, Karrer

, Landthaler

and Babilas

, Oxygen in acute and chronic wound healing, Br J Dermatol 163(2) (2010), 257–268.

Tandara

A.A.

and Mustoe

T.A.

, Oxygen in wound healing–more than a nutrient, World J Surg 28(3) (2004), 294–300.

10.

Sander

A.L.

, Henrich

, Muth

C.M.

, Marzi

, Barker

J.H.

and Frank

J.M.

, In vivo effect of hyperbaric oxygen on wound angiogenesis and epithelialization, Wound Repair Regen (2) (2009), 179–184.

11.

Forst

, Hohberg

, Tarakci

, Forst

, Kann

and Pfützner

, Reliability of lightguide spectrophotometry (O2C) for the investigation of skin tissue microvascular blood flow and tissue oxygen supply in diabetic and nondiabetic subjects, J Diabetes Sci Technol 2(6) (2008), 1151–1156.

12.

Dhar

S.C.

and Taylor

G.I.

, The delay phenomenon: The story unfolds, Plast Reconstr Surg 104(7) (1999), 2079–2091.

13.

Hamilton

, Wolfswinkel

E.M.

, Weathers

W.M.

, Xue

A.S.

, Hatef

D.A.

, Izaddoost

and Hollier

L.H.

Jr , The delay phenomenon: A compilation of knowledge across specialties, Craniomaxillofac Trauma Reconstr 7(2) (2014), 112–118.

14.

Knobloch

, Lichtenberg

, Pichlmaier

, Mertsching

, Krug

, Klima

and Haverich

, Microcirculation of the sternum following harvesting of the left internal mammary artery, Thorac Cardiovasc Surg 51(5) (2003), 255–259.

15.

Cederna

P.S.

, Chang

, Pittet-Cuenod

B.M.

, Razaboni

R.M.

and Cram

A.E.

, The effect of the delay phenomenon on the vascularity of rabbit rectus abdominis muscles, Plast Reconstr Surg 99(1) (1997), 194–205.

16.

Francel

T.J.

, Dufresne

C.R.

, Baumgartner

W.A.

and O’Kelley

, Anatomic and clinical considerations of an internal mammary artery harvest, Arch Surg 127(9) (1992), 1107–1111.

17.

Salehi Omran

, Karimi

, Ahmadi

S.H.

, Davoodi

, Marzban

, Movahedi

, Abbasi

, Boroumand

M.A.

, Davoodi

and Moshtaghi

, Superficial and deep sternal wound infection after more than 9000 coronary artery bypass graft (CABG): Incidence, risk factors and mortality, BMC Infect Dis 7 (2007), 112.

18.

Beckert

, Witte

M.B.

, Königsrainer

and Coerper

, The impact of the Micro-Lightguide O2C for the quantification of tissue ischemia in diabetic foot ulcers, Diabetes Care 27(12) (2004), 2863–2867.

19.

Taeger

C.D.

, Horch

R.E.

, Arkudas

, Schmitz

, Stübinger

, Lang

, Meyer

, Seitz

, Weyand

and Beier

J.P.

, Combined free flaps with arteriovenous loops for reconstruction of extensive thoracic defects after sternal osteomyelitis, Microsurgery 36(2) (2016), 121–127.

20.

Beier

J.P.

, Arkudas

, Lang

, Weyand

and Horch

R.E.

, Sternal osteomyelitis - Surgical treatment concepts, Chirurg 87(6) (2016), 537–550.

21.

Ried

, Geis

, Potzger

, Neu

, Klein

, Prantl

, Hofmann

H.S.

and Dolderer

J.H.

, Surgical reconstructive procedures of the chest wall after mediastinitis, Chirurg 87(6) (2016), 489–496.

22.

Liu

, Wang

, Yin

, Du

, Liu

, Yang

, Hu

, Wang

and Chen

, The stem cell adjuvant with Exendin-4 repairs the heart after myocardial infarction via STAT3 activation, J Cell Mol Med 18(7) (2014), 1381–1391.

23.

Horch

R.E.

, Beier

J.P.

, Kneser

and Arkudas

, Successful human long-term application of in situ bone tissue engineering, J Cell Mol Med 18(7) (2014), 1478–1485.

24.

Ranieri

, Klein

, Kotrade

, Nerlich

, Dolderer

, Prantl

and Geis

, Transepidermal oxygen flux measurement - First clinical application for postoperative wound monitoring, Clin Hemorheol Microcirc (2017). doi: 10.3233/CH-170265 [Epub ahead of print].

Patient characteristics
	Intervention	Control	P
Cohort size	n = 30	n = 30
Sex (male/female)	23/7	17/13	0.10
Mean age (Years)	66	62.4	0.21
Type of surgery	Coronary LIMA	Median sternotomy without IMA
	bypass (n = 30)	harvesting
		Aortic value reconstruction (n = 20)
		Mitral value reconstruction (n = 3)
		Tricuspid value reconstruction (n = 1)
		Ascending aorts reconstrucion (n = 5)
		Atrial myxoma removal (n = 1)
Diabetes mellitus	n = 9	n = 9	1.00
Hyperlipidemia	n = 18	n = 5	0.00
Hypertension	n = 23	n = 18	0.60
BMI >30	n = 10	n = 9	0.79
Alkohol	n = 16	n = 14	0.61
Nicotine	n = 6	n = 7	0.38