EurJ VascSurg 5, 571-576(1991)
The Specificity of Technetium-labelled-leucocyte Imaging of Aortic Grafts in the Early Postoperative Period R. L. Insall, P. M. Keavey 1, T. Hawkins 1, N. Hayes, N. A. G. Jones and J. Chamberlain Departments of Surgery and i Regional Medical Physics, Freeman Hospital, Newcastle-upon-Tyne, U.K. Radio-labelled-leucocyte imaging is becoming accepted as the invest(cation of choice when infection of prosthetic arterial bypass grafts is suspected. However, alternative sources of inftammation, such as haematoma and, perhaps, the healing and incorporation of newly inserted bypass grafts, are potential causes offalse positive results. This study examines the use of technetium-labelledleucocyte imaging during the postoperative in-hospital periodfollowing aortic bypass surgery. Two labelled-leucocytescans were carried out serially in 20 patients randomly allocated to receiveeither of two types of Dacron aorticgraft in regular use. The earlier scans were undertaken at 2-5 days after surgery and only two of these scans were positive. None of the later scans, undertaken at 7-10 days after surgery, was positive. None of the patients had a graft infection as evidenced by clinicalfeatures, leucocytecounts, and C-reactiveprotein measurements. These results suggest that recent surgery does not in itself signifieantly reduce the specificity of technetium-labelled-leucocyte imaging. Key Words: Arterial graft infection; Labelled-leucocyteimaging; Technetium hexametazime.
small number of patients with graft infection. Its accuracy, therefore, has not yet been fully established. Computerised tomography has relatively low sensiIn vascular surgery, infection of prosthetic arterial tivity. 11 Magnetic resonance imaging is not yet grafts is one of the more serious complications since, widely available, consequently experience of its use although it is uncommon, it leads to significant mor- in arterial graft infection is very limited and its accubidity, loss of limbs, and mortality in a high pro- racy remains to be determined. portion of cases. 1-3 In addition, graft infection usually Experience with 111indium-labelledqeucocytes necessitates removal of the graft and often revascular- has shown that inaccurate results may occur with low isation by an alternative means, commonly a difficult grade infection, platelet contamination, false and inventure. Treatment of graft infection is, therefore, a flammatory aneurysms, haematoma, acute inflammamajor undertaking although successful less radical tory conditions, enteric anastomoses, renal allografts, treatment has occasionally been reported. 4 The clini- and accessory splenic tissue. 6's' 12-14 In addition, up cal features of graft infection are non-specific and var- to 50% of arterial bypass graft patients may have posiied and the timing of presentation in relation to graft tive mindium-labelled-leucocyte scans in the early insertion may occur from the immediate post- postoperative period, is considerably reducing the operative period to many years later, s" 6 A high index value of this investigation for graft infection in this of suspicion and sensitive investigations are therefore important group of patients. The healing processes, required to establish the diagnosis and allow appro- involving leucocytes, during incorporation of a newly priate treatment. inserted prosthetic graft are proposed as the reason Recently, radiolabelled-leucocyte imaging has for these false positive results. become established as the investigation of choice for This study was undertaken to determine graft infection. 2'3"5-10 In particular, the n e w isotopic whether Tc-HMPAO-labelled-leucocytes share with label of technetium (99Tcm)-hexametazime (Tc- mindium-labelled-leucocyte imaging this problem of HMPAO, Amersham International PLC, Aylesbury, falsely positive results in the early postoperative UK) has several advantages over other current alter- period. Since intra-abdominal graft infection is the natives 5,10 but has so far been applied to a relatively site which is not only most dangerous but also most difficult to diagnose 1~ (relative to limb-graft infecPlea~e address all correspondence to: R. L. Insall, Department of tion), the study has been confined to patients underSurgery, Freeman Hospital, High Heaton, Newcastle-upon-Tyne, going aortic bypass surgery. NE7 7DN, U.K. There is a wide choice of graft materials available
Introduction
0950-821X/91/050571+06$03.00/0© 1991Grune & Stratton Ltd.
572
R.L. Insall etal.
for a o r t i c b y p a s s b u t D a c r o n w a s c h o s e n s i n c e it is t h e s t a n d a r d m a t e r i a l i n u s e in o u r u n i t a n d b e c a u s e it w a s a l s o u s e d b y S e d w i t z et al. in t h e e a r l i e r s t u d y o n 111indium-labelled-leucocyte i m a g i n g . 15 Woven velour or impregnated (chemically sealed) knitted Dacron grafts have largely replaced the earlier u n s e a l e d p o r o u s k n i t t e d grafts. A n e x a m p l e of b o t h t y p e s w a s s t u d i e d s i n c e t h e r e is a t h e o r e t i c a l p o s s i bility that graft impregnation might alter the tissue r e a c t i o n to t h e n e w l y i n s e r t e d graft a n d t h u s i n f l u e n c e t h e r e s u l t of a l a b e l l e d - l e u c o c y t e s c a n .
P a t i e n t s and M e t h o d s T w e n t y c o n s e c u t i v e p a t i e n t s , a d m i t t e d s e l e c t i v e l y for a o r t i c b y p a s s g r a f t s u r g e r y , c o n s e n t e d to p a r t i c i p a t e in t h i s s t u d y w h i c h w a s p e r m i t t e d b y t h e e t h i c a l c o m -
m i t t e e for t h e H e a l t h A u t h o r i t y . T h e p a t i e n t s w e r e a g e d b e t w e e n 52 a n d 81 y e a r s ( m e d i a n 65 y e a r s ) a n d t h e r e w e r e six w o m e n a n d 14 m e n . P a r t i c i p a t i o n i n the study did not influence the routine management of the patients. Straight aorto-aortic tube grafts were i n s e r t e d in 16 p a t i e n t s a n d f o u r h a d b i f u r c a t e d a o r t o b i f e m o r a l o r a o r t o - i l i a c grafts. F o u r p a t i e n t s h a d a o r t o - i l i a c o c c l u s i v e d i s e a s e a n d 16 p a t i e n t s h a d n o n inflammatory atherosclerotic abdominal aortic aneurysms. Patients were prospectively allocated by r a n d o m n u m b e r to r e c e i v e o n e of t w o t y p e s of graft in r e g u l a r u s e in o u r p r a c t i c e , V e r i s o f t o r H e m a s h i e l d D a c r o n g r a f t s ( M e a d o x U K L i m i t e d , D u n s t a b l e , UK). Routine standard anti-infective measures were taken, including intravenous antibiotic prophylaxis using a s i n g l e d o s e of g e n t a m i c i n 120 m g a n d f o u r d o s e s of f l u c l o x a c i l l i n 500 m g . Tc-HMPAO imaging was carried out twice d u r i n g t h e n o r m a l 1 0 - d a y p o s t o p e r a t i v e s t a y in all b u t
Table 1. Clinical details of the patients in the study
No.
Graft type
Graft site
C-reactive protein Day 1 Day 6
Home
Leucocyte count Preop Day 1
Postoperative course Day 6
Home
1
Verisoft
Aorto-aortic
115
62
48
8.6
8.8
5.9
8.7
Uneventful
2
Hemashield
Aortofemorat
138
89
67
6.5
6.4
12.3
8.1
Myocardial infarct
3
Verisoft
Aorto-aortic
114
91
85
3.9
14.4
4.5
4.8
Uneventful
4
Verisoft
Aorto-aortic
76
64
39
8.2
10.1
10.9
7.7
Uneventful
5
Hemashield
Aorto-iliac
120
102
63
10.2
10.4
15.4
8.2
Temporary diarrhoea
6
Hemashield
Aorto-aortic
167
158
70
10.0
13.6
14.7
10.2
Uneventful
7
Verisoft
Aorto-iliac
63
40
17
8.8
10.6
7.1
8.1
Uneventful
8
Hemashield
Aorto-aortic
59
31
22
7.6
8.2
8.1
7.5
Uneventful
9
Verisoft
Aorto-aortic
82
71
53
5.2
5.7
9.0
6.1
Uneventful
10
Hemashield
Aorto-aortic
94
74
57
7.4
7.9
12.4
6.6
Uneventful
11
Hemashield
Aorto-aortic
213
81
83
9.8
10.0
10.5
8.8
Uneventful
12
Verisoft
Aorto-aortic
262
119
29
7.6
8.9
12.2
8.0
Chest infection
13
Verisoft
Aorto-iliac
77
32
10
11.4
11.6
14.7
8.2
Uneventful
14
Hemashield
Aorto-aortic
60
47
36
8.6
8.6
12.5
7.8
Uneventful
15
Verisoft
Aorto-aortic
84
84
82
5.3
11.1
6.9
5.9
Uneventful
16
Hemashield
Aortofemoral
89
90
42
9.6
9.4
9.6
8.0
Uneventful
17
Hemashield
Aorto-aortic
120
83
46
7.2
7.8
10.1
8.0
Uneventful
18
Verisoft
Aorto-aortic
58
39
35
5.8
11.5
11.0
9.3
Uneventful
19
Verisoft
Aorto-aortic
142
69
43
9.6
10.1
6.7
8.3
Uneventful
20
Hemashield
Aorto-aortic
96
57
50
10.6
15.4
13.5
10.2
Uneventful
Eur J Vasc Surg Vol 5, October 1991
Technetium-labelled-leucocyte Imaging
two patients. The first scan was undertaken between 2 and 5 days when the patients had left intensive care and were sufficiently comfortable to attend for imaging. The second scan was carried out between 8 and 10 days after surgery, prior to the patients' discharge. One patient was discharged before the second scan and one patient agreed to return for a third scan 20 days after surgery, following earlier equivocal results. The method of Tc-HMPAO imaging was based on that of Danpure 16 and has been reported previously. 5 The scans were interpreted independently by two experienced observers with no detailed knowledge of the study patients. The scans were graded as positive (+), negative (-), or equivocal ( + / - ) . A scan was positive if activity in the region of the graft was visually greater than that of adjacent vascular features. Scans in which a confident assessment could not be made because of bowel activity overlying the area of the graft were graded as equivocal. The postoperative course of the patients was carefully monitored for potential signs suggesting infection, such as unexplained malaise or pyrexia. In addition, serial measurements of the leucocyte count and of C-reactive protein were carried out on days 1 and 6, and on the day of discharge from hospital. Details of the patients are summarised in Table 1. All of the patients have been reviewed at 6 months postoperatively, with no suggestion of graft infections detected.
Results All of the patients who were entered completed the study. No patient had clinical or haematological evidence suggesting graft infection. The postoperative serial leucocyte counts, expressed as a percentage of the preoperative value, and the C-reactive protein values are summarised in Table 1. These results support the absence of early graft infection in these patients. Initial Tc-HMPAO imaging was negative in 14 patients, positive in two, and equivocal in four patients. The second Tc-HMPAO scan was negative in 18 patients, positive in none, and equivocal in one patient. One patient was not studied due to early discharge from hospital before the second study could be undertaken; the first Tc-HMPAO scan was negative in this patient. One patient had a third study after equivocal results from the first two scans, this gave a negative result. Figure 1 shows an example of a positive scan with, for comparison, a normal or negative scan.
573
Thus, from a total of 40 scans, 33 were negative, five were equivocal (later negative), and two gave unexpected "false" positive results for the first scan. Tc-HMPAO imaging in this study, therefore, had an overall specificity of 82.5% if the equivocal results are included with the false positives. The specificity of the later scans, up to 10 days after surgery, was 95%. All of the patients had negative scans within 20 days of surgery. The timing distribution and results of the Tc-HMPAO scans are shown in Figure 2. No differences in any of the results were found between the two types of arterial graft used, Verisoft or Hemashield.
Discussion Previous reports have confirmed that Tc-HMPAOlabelled-leucocyte imaging is a highly sensitive predictor of arterial graft infection. 5"~° The accuracy of isotopic methods in general exceeds 90% in the published literature s" 6, 9-12,15 and this study does not set out to reassess the sensitivy of labelled-leucocyte imaging. Nevertheless, all but two positive and five equivocal Tc-HMPAO scans were negative in this study, an overall specificity of 82.5%. Since no patients with known graft infection were included in this study, the sensitivity of Tc-HMPAOqabelled-leucocyte imaging during this early postoperative period remains untested. Although not ideal markers, the leucocyte counts and C-reactive protein estimations are compatible with the absence of infection in these patients. This result contrasts with the poor specificity of 47% reported by Sedwitz et al. for mindium-labelledleucocyte imaging in the early period after graft insertion. ~s None of the four patients in this series with grafts to the groin had positive Tc-HMPAO scans, again at variance with the reported experience with 111indium-labelled-leucocytes.15 Moreover, in this series, no Tc-HMPAO scan was positive later than 5 days after surgery (though one patient had an equivocal result at 9 days) whereas mindium-labelledleucocyte positivity may be prolonged for over a year. 15 Even in the two scans reported as positive, activity in the region of the graft was not intense when compared to that in patients with confirmed graft infection and may simply represent a relatively small number of labelled leucocytes trapped in fibrin in and around the graft. Since the published specificities and sensitivities for Tc-HMPAO and mindium-labelled-leucocyte imaging overall are virtually identical, 5-12"14, 15 these Eur J VascSurg Vol 5, October1991
574
R.L. Insall eta/.
:m
Fig. 1. Examples of: a) positive (abnormal activity arrowed), and b) negative Tc-HMPAO images in patients with newl~c-implanted aortic bypass grafts. Eur J Vasc Surg Vol 5, October 1991
Technetium-labelled-leucocyte Imaging
575
12
I0
8 0
I
8
i
~2 'S E Z
0
t
0
0
0
[]
0
[]
0
[]
0
[]
0
[]
0
0
[]
O
@
o
[]
[]
(~
0
0
[]
[]
0
(~
0
0
[]
[]
•
•
o
@
[]
@
[]
z%
i 2
I 3
I 4
I 5
I 8
i 9
] I0
I 20
Doy of surgery
Pos,operotive cloys
Fig. 2. Scatter plot of times and results of serial Tc-HMPAO images in 20 patients with newly-implanted aortic bypass grafts. O, first scan negative; @, first scan positive; @, first scan equivocal; [3, second scan negative; @, second scan equivocal; A, third scan negative.
differences in specificity in the early phase after graft insertion are not readily explicable. It is possible that differences in surgical technique or graft materials may have contributed to this variance. Non-impregnated knitted Dacron grafts (Vasculour II, Bard USCI) were largely used by Sedwitz et al. ~s compared to the low porosity woven Dacron (Verisoft) or collagenimpregnated knitted Dacron (Hemashield) grafts used in this study. The high porosity of non-impregnated knitted Dacron requires preclotting but ensures earlier formation of a secure neo-intima and "healing" or incorporation by fibrous tissue, usually within 3 months. 17 In this study, no difference between the knitted and woven grafts was found. These findings argue against an effect due to the type of grafts used. No comment regarding surgical technique can be made since the precise operative methods employed by the Sedwitz group were not reported is and are unknown to us. The results of this study suggest that TcHMPAO-labelled-leucocyte imaging can be used with acceptable specificity in the immediate postoperative period following aortic bypass grafting. Tc-HMPAOlabelled-leucocytes do not share the low specificity of 111indium-labelled-leucocyte imaging during this important period. This may represent a further significant advantage in favour of Tc-HMPAO. However, this work requires confirmation and, as yet, no direct comparison between the two isotopic techniques has been reported. Nevertheless, when considered in conjunction with the several other advantages reported in favour of Tc-HMPAO relative to other current techniques, s" 10,13 Tc-HMPAO-labelled-leuco-
cyte imaging can again be recommended as the first choice investigation w h e n arterial graft infection is suspected.
Acknowledgements
We are grateful to the staff of the Regional Medical Physics Department who carried out the imaging of patients and to Meadox UK and Amersham International for financial support.
References 1 BUNT TJ. Synthetic vascular graft infections. Surgery 1983; 93: 733-746. 2 FREISCHLAG JA, MOORE WS. Infection in prosthetic arterial grafts. In RUTHERFORDRB, ed. Vascular Surgery. Philadelphia: WB Saunders Company, 1989; 40: 510-521. 3 BANDYKDF. Graft infection: a dreadful challenge. Seminars in Vascular Surgery 1990; 3: 77-80. 4 BAII~E¥IS, BUNDREDNJ, PEARSONHJ, BELLPRF. Successful treatrnent of an infected vascular graft with Gentamicin beads. Eur J Vasc Surg 1987; 1: 143-144. 5 INSALL RL, JONES NAG, CHAMBERLAIN J, LAMBERT D, KEAVEY PM. A new isotopic technique for detecting prosthetic arterial graft infection: 99"Tcrn-Hexametazime-labelled-leucocyte imaging. Br J Surg 1990; 77: 1295-1298.
6 BRUNNER,MC, MITCHELLRS, BALDWINJC,JAMEsDR, OLCOTTC, MEHIGANJT, McDOUGALLIR, MILLERDC. Prosthetic graft infection: limitations of indium white blood cell scanning. J Vasc Surg 1986; 3: 42-48. 7 BERRIDGEDC, EARNSHAWJJ, FRIERM, PERKINSAC, WASTIEML, HOPKINSON BR, MAKINGS. 111In-labelled leucocyte imaging in vascular graft infection. Br J Sur£ 1989; 76: 41-44. Eur J Vasc Surg Vol 5, October 1991
576
R.L. Insall etaL
8 REILLYDT, GRIGGMJ, CUNNINGHAMDA, THOMASEJ, MANSFIELD AO. Vascular graft infection: the role of indium scanning. Eur J Vasc Surg 1989; 3: 393-397. 9 WILLIAMSONMR, BOYD, CM, READ RC, THOMPSON BW, BARNES RW, SHAH HR, BALACHANDRAN S, FERRISEJ. 111in_labelle d leucocytes in the detection of prosthetic vascular graft infections. Am J Roentgenol 1986; 147: 173-176. 10 VORNE M, SOINI I, LANTTO T, PAAKKINEN S. Technetium-99m HMPAO-labelled-leucocytes in detection of inflammatory lesions: comparison with Gallium-67 citrate. J Nucl Med 1989; 30: 1332-1336. 11 ~A]ILLIAMSONMR, BOYD CM, READRC, et al. 111In-labelled leucocytes in the detection of prosthetic vascular graft infections. AJR 1986; 147: 173-176. 12 McAFEE JG, SAMIN A. In-111 labelled leucocytes: A review of problems in image interpretation. Radiology 1985; 155: 221-229.
Eur J Vasc Surg Vol 5, October 1991
13 PETERS AM. Sepsis (New approaches to imaging and thrombus used at St. Bartholomew's Hospital 1988). Nucl Med Commun 1988; 9: 719-723. 14 MERRELLSW, LAWRENCEPF. Diagnosis of graft infection: anatomic and functional imaging techniques. Seminars in Vascular Surgery 1990; 3: 89-100. 15 Sedwitz MM, Davies RJ, Pretorius HT, Vasquez TE. indium 111-labelled white blood cell scans after vascular prosthetic reconstruction. J Vasc Surg 1987; 6: 476-481. 16 DANPUREHJ, OSMAN S, CARROLLMK. In vitro studies to develop a clinical protocol for radioqabelling mixed leucocytes with 99Tcm-HMPAO. Nucl Med Commun 1987; 8: 280. 17 SMITH SRG. The principles underlying graft selection for arterial reconstruction. Br J Hosp Med 1987; 38: 358-363.
Accepted 16 April 1991
The Specificity of Technetium-labelled-leucocyte Imaging of Aortic Grafts in the Early Postoperative Period R. L. Insall, P. M. Keavey 1, T. Hawkins 1, N. Hayes, N. A. G. Jones and J. Chamberlain Departments of Surgery and i Regional Medical Physics, Freeman Hospital, Newcastle-upon-Tyne, U.K. Radio-labelled-leucocyte imaging is becoming accepted as the invest(cation of choice when infection of prosthetic arterial bypass grafts is suspected. However, alternative sources of inftammation, such as haematoma and, perhaps, the healing and incorporation of newly inserted bypass grafts, are potential causes offalse positive results. This study examines the use of technetium-labelledleucocyte imaging during the postoperative in-hospital periodfollowing aortic bypass surgery. Two labelled-leucocytescans were carried out serially in 20 patients randomly allocated to receiveeither of two types of Dacron aorticgraft in regular use. The earlier scans were undertaken at 2-5 days after surgery and only two of these scans were positive. None of the later scans, undertaken at 7-10 days after surgery, was positive. None of the patients had a graft infection as evidenced by clinicalfeatures, leucocytecounts, and C-reactiveprotein measurements. These results suggest that recent surgery does not in itself signifieantly reduce the specificity of technetium-labelled-leucocyte imaging. Key Words: Arterial graft infection; Labelled-leucocyteimaging; Technetium hexametazime.
small number of patients with graft infection. Its accuracy, therefore, has not yet been fully established. Computerised tomography has relatively low sensiIn vascular surgery, infection of prosthetic arterial tivity. 11 Magnetic resonance imaging is not yet grafts is one of the more serious complications since, widely available, consequently experience of its use although it is uncommon, it leads to significant mor- in arterial graft infection is very limited and its accubidity, loss of limbs, and mortality in a high pro- racy remains to be determined. portion of cases. 1-3 In addition, graft infection usually Experience with 111indium-labelledqeucocytes necessitates removal of the graft and often revascular- has shown that inaccurate results may occur with low isation by an alternative means, commonly a difficult grade infection, platelet contamination, false and inventure. Treatment of graft infection is, therefore, a flammatory aneurysms, haematoma, acute inflammamajor undertaking although successful less radical tory conditions, enteric anastomoses, renal allografts, treatment has occasionally been reported. 4 The clini- and accessory splenic tissue. 6's' 12-14 In addition, up cal features of graft infection are non-specific and var- to 50% of arterial bypass graft patients may have posiied and the timing of presentation in relation to graft tive mindium-labelled-leucocyte scans in the early insertion may occur from the immediate post- postoperative period, is considerably reducing the operative period to many years later, s" 6 A high index value of this investigation for graft infection in this of suspicion and sensitive investigations are therefore important group of patients. The healing processes, required to establish the diagnosis and allow appro- involving leucocytes, during incorporation of a newly priate treatment. inserted prosthetic graft are proposed as the reason Recently, radiolabelled-leucocyte imaging has for these false positive results. become established as the investigation of choice for This study was undertaken to determine graft infection. 2'3"5-10 In particular, the n e w isotopic whether Tc-HMPAO-labelled-leucocytes share with label of technetium (99Tcm)-hexametazime (Tc- mindium-labelled-leucocyte imaging this problem of HMPAO, Amersham International PLC, Aylesbury, falsely positive results in the early postoperative UK) has several advantages over other current alter- period. Since intra-abdominal graft infection is the natives 5,10 but has so far been applied to a relatively site which is not only most dangerous but also most difficult to diagnose 1~ (relative to limb-graft infecPlea~e address all correspondence to: R. L. Insall, Department of tion), the study has been confined to patients underSurgery, Freeman Hospital, High Heaton, Newcastle-upon-Tyne, going aortic bypass surgery. NE7 7DN, U.K. There is a wide choice of graft materials available
Introduction
0950-821X/91/050571+06$03.00/0© 1991Grune & Stratton Ltd.
572
R.L. Insall etal.
for a o r t i c b y p a s s b u t D a c r o n w a s c h o s e n s i n c e it is t h e s t a n d a r d m a t e r i a l i n u s e in o u r u n i t a n d b e c a u s e it w a s a l s o u s e d b y S e d w i t z et al. in t h e e a r l i e r s t u d y o n 111indium-labelled-leucocyte i m a g i n g . 15 Woven velour or impregnated (chemically sealed) knitted Dacron grafts have largely replaced the earlier u n s e a l e d p o r o u s k n i t t e d grafts. A n e x a m p l e of b o t h t y p e s w a s s t u d i e d s i n c e t h e r e is a t h e o r e t i c a l p o s s i bility that graft impregnation might alter the tissue r e a c t i o n to t h e n e w l y i n s e r t e d graft a n d t h u s i n f l u e n c e t h e r e s u l t of a l a b e l l e d - l e u c o c y t e s c a n .
P a t i e n t s and M e t h o d s T w e n t y c o n s e c u t i v e p a t i e n t s , a d m i t t e d s e l e c t i v e l y for a o r t i c b y p a s s g r a f t s u r g e r y , c o n s e n t e d to p a r t i c i p a t e in t h i s s t u d y w h i c h w a s p e r m i t t e d b y t h e e t h i c a l c o m -
m i t t e e for t h e H e a l t h A u t h o r i t y . T h e p a t i e n t s w e r e a g e d b e t w e e n 52 a n d 81 y e a r s ( m e d i a n 65 y e a r s ) a n d t h e r e w e r e six w o m e n a n d 14 m e n . P a r t i c i p a t i o n i n the study did not influence the routine management of the patients. Straight aorto-aortic tube grafts were i n s e r t e d in 16 p a t i e n t s a n d f o u r h a d b i f u r c a t e d a o r t o b i f e m o r a l o r a o r t o - i l i a c grafts. F o u r p a t i e n t s h a d a o r t o - i l i a c o c c l u s i v e d i s e a s e a n d 16 p a t i e n t s h a d n o n inflammatory atherosclerotic abdominal aortic aneurysms. Patients were prospectively allocated by r a n d o m n u m b e r to r e c e i v e o n e of t w o t y p e s of graft in r e g u l a r u s e in o u r p r a c t i c e , V e r i s o f t o r H e m a s h i e l d D a c r o n g r a f t s ( M e a d o x U K L i m i t e d , D u n s t a b l e , UK). Routine standard anti-infective measures were taken, including intravenous antibiotic prophylaxis using a s i n g l e d o s e of g e n t a m i c i n 120 m g a n d f o u r d o s e s of f l u c l o x a c i l l i n 500 m g . Tc-HMPAO imaging was carried out twice d u r i n g t h e n o r m a l 1 0 - d a y p o s t o p e r a t i v e s t a y in all b u t
Table 1. Clinical details of the patients in the study
No.
Graft type
Graft site
C-reactive protein Day 1 Day 6
Home
Leucocyte count Preop Day 1
Postoperative course Day 6
Home
1
Verisoft
Aorto-aortic
115
62
48
8.6
8.8
5.9
8.7
Uneventful
2
Hemashield
Aortofemorat
138
89
67
6.5
6.4
12.3
8.1
Myocardial infarct
3
Verisoft
Aorto-aortic
114
91
85
3.9
14.4
4.5
4.8
Uneventful
4
Verisoft
Aorto-aortic
76
64
39
8.2
10.1
10.9
7.7
Uneventful
5
Hemashield
Aorto-iliac
120
102
63
10.2
10.4
15.4
8.2
Temporary diarrhoea
6
Hemashield
Aorto-aortic
167
158
70
10.0
13.6
14.7
10.2
Uneventful
7
Verisoft
Aorto-iliac
63
40
17
8.8
10.6
7.1
8.1
Uneventful
8
Hemashield
Aorto-aortic
59
31
22
7.6
8.2
8.1
7.5
Uneventful
9
Verisoft
Aorto-aortic
82
71
53
5.2
5.7
9.0
6.1
Uneventful
10
Hemashield
Aorto-aortic
94
74
57
7.4
7.9
12.4
6.6
Uneventful
11
Hemashield
Aorto-aortic
213
81
83
9.8
10.0
10.5
8.8
Uneventful
12
Verisoft
Aorto-aortic
262
119
29
7.6
8.9
12.2
8.0
Chest infection
13
Verisoft
Aorto-iliac
77
32
10
11.4
11.6
14.7
8.2
Uneventful
14
Hemashield
Aorto-aortic
60
47
36
8.6
8.6
12.5
7.8
Uneventful
15
Verisoft
Aorto-aortic
84
84
82
5.3
11.1
6.9
5.9
Uneventful
16
Hemashield
Aortofemoral
89
90
42
9.6
9.4
9.6
8.0
Uneventful
17
Hemashield
Aorto-aortic
120
83
46
7.2
7.8
10.1
8.0
Uneventful
18
Verisoft
Aorto-aortic
58
39
35
5.8
11.5
11.0
9.3
Uneventful
19
Verisoft
Aorto-aortic
142
69
43
9.6
10.1
6.7
8.3
Uneventful
20
Hemashield
Aorto-aortic
96
57
50
10.6
15.4
13.5
10.2
Uneventful
Eur J Vasc Surg Vol 5, October 1991
Technetium-labelled-leucocyte Imaging
two patients. The first scan was undertaken between 2 and 5 days when the patients had left intensive care and were sufficiently comfortable to attend for imaging. The second scan was carried out between 8 and 10 days after surgery, prior to the patients' discharge. One patient was discharged before the second scan and one patient agreed to return for a third scan 20 days after surgery, following earlier equivocal results. The method of Tc-HMPAO imaging was based on that of Danpure 16 and has been reported previously. 5 The scans were interpreted independently by two experienced observers with no detailed knowledge of the study patients. The scans were graded as positive (+), negative (-), or equivocal ( + / - ) . A scan was positive if activity in the region of the graft was visually greater than that of adjacent vascular features. Scans in which a confident assessment could not be made because of bowel activity overlying the area of the graft were graded as equivocal. The postoperative course of the patients was carefully monitored for potential signs suggesting infection, such as unexplained malaise or pyrexia. In addition, serial measurements of the leucocyte count and of C-reactive protein were carried out on days 1 and 6, and on the day of discharge from hospital. Details of the patients are summarised in Table 1. All of the patients have been reviewed at 6 months postoperatively, with no suggestion of graft infections detected.
Results All of the patients who were entered completed the study. No patient had clinical or haematological evidence suggesting graft infection. The postoperative serial leucocyte counts, expressed as a percentage of the preoperative value, and the C-reactive protein values are summarised in Table 1. These results support the absence of early graft infection in these patients. Initial Tc-HMPAO imaging was negative in 14 patients, positive in two, and equivocal in four patients. The second Tc-HMPAO scan was negative in 18 patients, positive in none, and equivocal in one patient. One patient was not studied due to early discharge from hospital before the second study could be undertaken; the first Tc-HMPAO scan was negative in this patient. One patient had a third study after equivocal results from the first two scans, this gave a negative result. Figure 1 shows an example of a positive scan with, for comparison, a normal or negative scan.
573
Thus, from a total of 40 scans, 33 were negative, five were equivocal (later negative), and two gave unexpected "false" positive results for the first scan. Tc-HMPAO imaging in this study, therefore, had an overall specificity of 82.5% if the equivocal results are included with the false positives. The specificity of the later scans, up to 10 days after surgery, was 95%. All of the patients had negative scans within 20 days of surgery. The timing distribution and results of the Tc-HMPAO scans are shown in Figure 2. No differences in any of the results were found between the two types of arterial graft used, Verisoft or Hemashield.
Discussion Previous reports have confirmed that Tc-HMPAOlabelled-leucocyte imaging is a highly sensitive predictor of arterial graft infection. 5"~° The accuracy of isotopic methods in general exceeds 90% in the published literature s" 6, 9-12,15 and this study does not set out to reassess the sensitivy of labelled-leucocyte imaging. Nevertheless, all but two positive and five equivocal Tc-HMPAO scans were negative in this study, an overall specificity of 82.5%. Since no patients with known graft infection were included in this study, the sensitivity of Tc-HMPAOqabelled-leucocyte imaging during this early postoperative period remains untested. Although not ideal markers, the leucocyte counts and C-reactive protein estimations are compatible with the absence of infection in these patients. This result contrasts with the poor specificity of 47% reported by Sedwitz et al. for mindium-labelledleucocyte imaging in the early period after graft insertion. ~s None of the four patients in this series with grafts to the groin had positive Tc-HMPAO scans, again at variance with the reported experience with 111indium-labelled-leucocytes.15 Moreover, in this series, no Tc-HMPAO scan was positive later than 5 days after surgery (though one patient had an equivocal result at 9 days) whereas mindium-labelledleucocyte positivity may be prolonged for over a year. 15 Even in the two scans reported as positive, activity in the region of the graft was not intense when compared to that in patients with confirmed graft infection and may simply represent a relatively small number of labelled leucocytes trapped in fibrin in and around the graft. Since the published specificities and sensitivities for Tc-HMPAO and mindium-labelled-leucocyte imaging overall are virtually identical, 5-12"14, 15 these Eur J VascSurg Vol 5, October1991
574
R.L. Insall eta/.
:m
Fig. 1. Examples of: a) positive (abnormal activity arrowed), and b) negative Tc-HMPAO images in patients with newl~c-implanted aortic bypass grafts. Eur J Vasc Surg Vol 5, October 1991
Technetium-labelled-leucocyte Imaging
575
12
I0
8 0
I
8
i
~2 'S E Z
0
t
0
0
0
[]
0
[]
0
[]
0
[]
0
[]
0
0
[]
O
@
o
[]
[]
(~
0
0
[]
[]
0
(~
0
0
[]
[]
•
•
o
@
[]
@
[]
z%
i 2
I 3
I 4
I 5
I 8
i 9
] I0
I 20
Doy of surgery
Pos,operotive cloys
Fig. 2. Scatter plot of times and results of serial Tc-HMPAO images in 20 patients with newly-implanted aortic bypass grafts. O, first scan negative; @, first scan positive; @, first scan equivocal; [3, second scan negative; @, second scan equivocal; A, third scan negative.
differences in specificity in the early phase after graft insertion are not readily explicable. It is possible that differences in surgical technique or graft materials may have contributed to this variance. Non-impregnated knitted Dacron grafts (Vasculour II, Bard USCI) were largely used by Sedwitz et al. ~s compared to the low porosity woven Dacron (Verisoft) or collagenimpregnated knitted Dacron (Hemashield) grafts used in this study. The high porosity of non-impregnated knitted Dacron requires preclotting but ensures earlier formation of a secure neo-intima and "healing" or incorporation by fibrous tissue, usually within 3 months. 17 In this study, no difference between the knitted and woven grafts was found. These findings argue against an effect due to the type of grafts used. No comment regarding surgical technique can be made since the precise operative methods employed by the Sedwitz group were not reported is and are unknown to us. The results of this study suggest that TcHMPAO-labelled-leucocyte imaging can be used with acceptable specificity in the immediate postoperative period following aortic bypass grafting. Tc-HMPAOlabelled-leucocytes do not share the low specificity of 111indium-labelled-leucocyte imaging during this important period. This may represent a further significant advantage in favour of Tc-HMPAO. However, this work requires confirmation and, as yet, no direct comparison between the two isotopic techniques has been reported. Nevertheless, when considered in conjunction with the several other advantages reported in favour of Tc-HMPAO relative to other current techniques, s" 10,13 Tc-HMPAO-labelled-leuco-
cyte imaging can again be recommended as the first choice investigation w h e n arterial graft infection is suspected.
Acknowledgements
We are grateful to the staff of the Regional Medical Physics Department who carried out the imaging of patients and to Meadox UK and Amersham International for financial support.
References 1 BUNT TJ. Synthetic vascular graft infections. Surgery 1983; 93: 733-746. 2 FREISCHLAG JA, MOORE WS. Infection in prosthetic arterial grafts. In RUTHERFORDRB, ed. Vascular Surgery. Philadelphia: WB Saunders Company, 1989; 40: 510-521. 3 BANDYKDF. Graft infection: a dreadful challenge. Seminars in Vascular Surgery 1990; 3: 77-80. 4 BAII~E¥IS, BUNDREDNJ, PEARSONHJ, BELLPRF. Successful treatrnent of an infected vascular graft with Gentamicin beads. Eur J Vasc Surg 1987; 1: 143-144. 5 INSALL RL, JONES NAG, CHAMBERLAIN J, LAMBERT D, KEAVEY PM. A new isotopic technique for detecting prosthetic arterial graft infection: 99"Tcrn-Hexametazime-labelled-leucocyte imaging. Br J Surg 1990; 77: 1295-1298.
6 BRUNNER,MC, MITCHELLRS, BALDWINJC,JAMEsDR, OLCOTTC, MEHIGANJT, McDOUGALLIR, MILLERDC. Prosthetic graft infection: limitations of indium white blood cell scanning. J Vasc Surg 1986; 3: 42-48. 7 BERRIDGEDC, EARNSHAWJJ, FRIERM, PERKINSAC, WASTIEML, HOPKINSON BR, MAKINGS. 111In-labelled leucocyte imaging in vascular graft infection. Br J Sur£ 1989; 76: 41-44. Eur J Vasc Surg Vol 5, October 1991
576
R.L. Insall etaL
8 REILLYDT, GRIGGMJ, CUNNINGHAMDA, THOMASEJ, MANSFIELD AO. Vascular graft infection: the role of indium scanning. Eur J Vasc Surg 1989; 3: 393-397. 9 WILLIAMSONMR, BOYD, CM, READ RC, THOMPSON BW, BARNES RW, SHAH HR, BALACHANDRAN S, FERRISEJ. 111in_labelle d leucocytes in the detection of prosthetic vascular graft infections. Am J Roentgenol 1986; 147: 173-176. 10 VORNE M, SOINI I, LANTTO T, PAAKKINEN S. Technetium-99m HMPAO-labelled-leucocytes in detection of inflammatory lesions: comparison with Gallium-67 citrate. J Nucl Med 1989; 30: 1332-1336. 11 ~A]ILLIAMSONMR, BOYD CM, READRC, et al. 111In-labelled leucocytes in the detection of prosthetic vascular graft infections. AJR 1986; 147: 173-176. 12 McAFEE JG, SAMIN A. In-111 labelled leucocytes: A review of problems in image interpretation. Radiology 1985; 155: 221-229.
Eur J Vasc Surg Vol 5, October 1991
13 PETERS AM. Sepsis (New approaches to imaging and thrombus used at St. Bartholomew's Hospital 1988). Nucl Med Commun 1988; 9: 719-723. 14 MERRELLSW, LAWRENCEPF. Diagnosis of graft infection: anatomic and functional imaging techniques. Seminars in Vascular Surgery 1990; 3: 89-100. 15 Sedwitz MM, Davies RJ, Pretorius HT, Vasquez TE. indium 111-labelled white blood cell scans after vascular prosthetic reconstruction. J Vasc Surg 1987; 6: 476-481. 16 DANPUREHJ, OSMAN S, CARROLLMK. In vitro studies to develop a clinical protocol for radioqabelling mixed leucocytes with 99Tcm-HMPAO. Nucl Med Commun 1987; 8: 280. 17 SMITH SRG. The principles underlying graft selection for arterial reconstruction. Br J Hosp Med 1987; 38: 358-363.
Accepted 16 April 1991
