13 Chronic transplant dysfunction

Introduction

The success of organ transplantation is marred by the fact that some initially successful transplants have suboptimal function or deteriorate over time. Despite advances in the control of acute rejection, the problem of chronic transplant dysfunction (CTD) remains significant.

CTD is characterised by a progressive deterioration in graft function over the months and years following transplantation, and is associated with characteristic histological features of fibrosis and graft arteriosclerosis. By the time of diagnosis, the changes that have taken place are usually irreversible and lead inexorably to graft loss. For the majority of solid organs transplanted, CTD is the leading cause of graft loss after the first year. At 5 years post-transplantation, 30–50% of kidney, heart, lung and pancreas, and 5–20% of liver allografts demonstrate typical morphological changes.

These data must be considered in conjunction with the consistent improvement in 1-year graft survival rates that have been achieved, due to more effective immunosuppressive therapy and recipient/donor care. The 1-year survival rate is 80–95% for kidney, liver, heart and lung transplantation, and so it is now necessary to turn our attention to the causes of late graft loss.

The corresponding chapter in the previous edition of this book provided an excellent overview of CTD.¹ Although a brief description of organ-specific CTD is provided, this chapter will primarily adopt renal transplantation as the paradigm for study of CTD. It will explore developments in our understanding of the condition, and expand upon immunological factors that influence this pathological entity. The adoption of protocol biopsies in some centres across the world has allowed the investigation of early changes in renal histology that lead to CTD, and so this concept will be discussed.

Organ-specific findings

After excluding organ-specific disease (such as hepatitis C or recurrent glomerulonephritis) characteristic but non-specific patterns of pathology can be described in different organs.

Heart

CTD in transplanted hearts remains the most common cause of graft loss after the first post-transplant year. It is manifested as accelerated cardiac allograft arteriosclerosis or cardiac allograft vasculopathy (CAV). The diagnosis of CAV relies on clinical and radiographic features of arteriosclerosis. The pathognomonic lesion of CAV in heart allografts is concentric intimal thickening affecting the coronary vessels, particularly the smaller distal intramyocardial vessels.

Diagnostic criteria for CAV were presented in the consensus document of 1993 from the fourth Alexis Carrel Conference.²

The document proposed the following guidelines:

1 Histological diagnosis should be based on the presence of tubular myointimal hyperplasia with an intact elastic lamina containing few breaks. There should be little or no calcification.

2 The histopathological diagnosis should be made on explanted transplants or at autopsy, recognising that endomyocardial biopsies rarely contain arteries.

3 The clinical diagnosis should be made on angiography and reported according to categories described by the Stanford group:³

a Type A lesions – discrete stenoses in the proximal, middle or distal segment branches.

b Type B lesions – diffuse concentric narrowing arising in the middle to distal arteries.

c Type C lesions – narrowed irregular vessels with occluded side branches.

Type A changes in the proximal arteries are likely to represent prior coronary artery disease or new-onset atherosclerosis typical of that found in native hearts.

Liver

Chronic liver rejection is again one of the most common causes of late graft loss, but it occurs far less frequently than in other solid-organ transplants. The overall incidence of chronic transplant dysfunction is between 5% and 20% of all liver transplants.³

The liver is immunologically privileged, with all forms of rejection being less common. Specifically, CTD is becoming less common,⁴ with graft loss more likely from recurrent disease, late acute rejection or non-compliance with immunosuppression, but it is still a major indication for re-transplantation.^5,⁶

Histological diagnosis is also essential, with liver cell dropout, vanishing bile duct syndrome and obliterative vasculopathy being the cardinal features.⁷ ‘Vanishing bile duct syndrome’ when diagnosed on biopsy is a good indicator for severe chronic allograft damage. It is a uniform loss of small bile ducts (<75 μm) throughout the liver⁸ without replacement by fibrosis. Ductule proliferation does not occur as in other cholestatic conditions, such as primary biliary cirrhosis. When seen in biopsy specimens, large and medium-sized arteries have a foam cell and macrophage-laden intima as is found in other allografted organs.

Lung

Bronchiolitis obliterans syndrome (BOS) is the clinical manifestation of CTD in the lung, and the typical histopathological features are referred to as obliterative bronchiolitis. BOS accounts for over one-third of late graft losses and the prevalence of BOS in patients greater than 3 months post-transplantation may be as high as 68%.⁹ The mortality rate is 50% once the diagnosis of BOS has been made, with few patients coming for re-transplantation. In common with the disease process in other organs, BOS is progressive and has no effective treatment at present. Clinically it presents with slowly progressing dyspnoea on exertion. Worsening airflow obstruction follows as a consequence of deteriorating graft function. Physiologically there is a mixed picture of airflow obstruction and restrictive pulmonary disease.¹⁰

Histologically there is inflammation and fibrosis of the cartilaginous airways and particularly within the smaller airways. Bronchioles usually demonstrate areas of inflammation and fibrosis in the lamina propria and luminal surfaces whilst the larger bronchi show peribronchial fibrosis and bronchiectasis. Airway narrowing follows, accounting for the decline in spirometry readings.¹¹ Surrounding alveoli and interstitium are often, but not always, normal. As a pathological entity it is no different from obliterative bronchiolitis caused by non-transplant-related aetiological factors, such as toxic fume inhalation, drug side-effects and connective tissue disorders.

In 1990 an ad hoc committee under the auspices of the International Society for Heart and Lung Transplantation proposed a working formulation for the standardisation of nomenclature and for clinical staging of chronic dysfunction in lung allografts.¹²

The formulation described the following staging system:

1 Stage 0 – no significant abnormality; forced expiratory volume in 1 second (FEV₁) 80% or more of baseline value.

2 Stage 1 – mild bronchiolitis obliterans syndrome; FEV₁ 66–88%.

3 Stage 2 – moderate bronchiolitis obliterans syndrome; FEV₁ 51–65%.

4 Stage 3 – severe bronchiolitis obliterans syndrome; FEV₁ 50% or less.

Within each stage, ‘a’ and ‘b’ categories exist: ‘a’ denotes no histological evidence of bronchiolitis obliterans or an absence of biopsy material and ‘b’ denotes a positive diagnosis obtained on biopsy material.

Pancreas

CTD accounts for over half of all pancreas graft losses 5 years post-transplantation.¹³ Deterioration in graft function is not easy to detect and is usually manifested by hyperglycaemic episodes with a low C-peptide level following a glucose challenge.¹⁴ There are no serial markers of graft deterioration in pancreas transplantation and, in the absence of concurrent acute rejection, serum amylase is often normal.

Graft biopsy is required to confirm the diagnosis. The typical histological features seen are septal fibrosis with acinar loss and fibrointimal vascular proliferation. Vessels are seldom seen on graft biopsies and therefore vascular changes cannot be relied on for diagnosis.

Thus it is observed that different organs deteriorate with different patterns, and these echo to some extent the patterns of ageing: the atrophic and fibrotic processes in kidney transplants resemble the much slower shutdown of nephrons in normal ageing, the ‘Achilles heel’ of the transplanted and native heart is coronary artery deterioration, and lung transplants develop increased small airway resistance, just as normal lungs do with age and environmental factors, such as smoking.

Renal transplantation: the paradigm for chronic allograft injury

Chronic renal allograft dysfunction remains one of the most common causes of graft loss beyond the first year, and has been the subject of study over recent years. The adoption of serial protocol biopsies in some centres has led to an improvement in our understanding of the process. The pattern of injury is similar to that seen in other organs, such as the heart and lungs, and so much of the work is relevant to all organs.

Protocol biopsies

Renal allograft biopsy is the most accurate method for identifying pathophysiological events that have a bearing on short- and long-term graft outcomes. Allograft biopsies are traditionally performed in the setting of acute or chronic deterioration in graft function. Recent years have seen increasing recognition that a measurable decrease in renal function does not always accompany acute rejection or chronic changes such as tubular atrophy or interstitial fibrosis, and this has given rise to the practice of performing biopsies on stable allografts at predefined post-transplant periods, known as protocol biopsies.

Protocol biopsies and subclinical rejection

Protocol biopsies allow the early detection of rejection, at a time before this is manifested clinically, known as subclinical rejection (SCR). SCR is defined as histological evidence of acute rejection in patients with stable renal function (less than 25% change in serum creatinine), or borderline changes, a term that is used when no intimal arteritis is present, but there are foci of tubulitis that do not meet criteria for rejection diagnosis, also in patients with stable renal function. Thus, SCR may be important as a precursor to clinically relevant rejection episodes, allowing early intervention. Moreover, persistent, unrecognised inflammatory injury, secondary to a low-grade rejection process, may result in long-term graft fibrosis and chronic dysfunction without overt clinical rejection, and this could be identified on protocol biopsies.¹⁵

In order to justify a protocol biopsy with its attendant risks, a clear link must be defined between SCR and subsequent development of chronic graft dysfunction. In one study, SCR diagnosed in early biopsies (1–3 months) was treated with corticosteroids, and this was associated with both a reduction in late acute rejection and also in chronic damage compared with controls.¹⁶ This has been supported in other studies, suggesting that the use of protocol biopsies for the early detection and treatment of subclinical rejection may be a major factor in preserving long-term graft function.^17,¹⁸

Protocol biopsies and chronic allograft changes

In addition, protocol biopsies may allow early detection of specific chronic allograft changes, such as tubular atrophy and interstitial fibrosis (TA/IF). Such chronic changes are a frequent finding in protocol biopsies, and are seen to progress rapidly over the first year and more slowly thereafter.^19,²⁰ Helantera et al. undertook a study to determine the optimal timing of protocol biopsies and used the chronic allograft damage index (CADI), defined at the Banff 1997 meeting to define chronic injury.²¹ Follow-up was for 18 months, and this demonstrated that CADI at 6 and 12 months was associated with long-term graft survival, but not that at 3 months.²² If the diagnosis of early changes of fibrosis and atrophy are detected on protocol biopsy, the real challenge is to identify effective therapeutic strategies leading to improved outcome. The diagnosis alone is unlikely to have a positive impact on the patient.

Protocol biopsies and non-immunological contributors to chronic graft dysfunction

Non-immunological factors also play a significant role in the subsequent development of chronic allograft dysfunction and these, having been identified early in protocol biopsies, may be amenable to specific treatment. One of the most common factors is calcineurin inhibitor (CNI) toxicity, which can be reversed by modifying immunosuppressive therapy. A small study showed that the histological change characteristic of CNI toxicity (hyaline arteriolar sclerosis), when seen in early protocol biopsies (<1 year), was an independent risk factor for subsequent development of chronic allograft nephropathy.²³ This supports earlier work by Nankivell et al., who undertook a longitudinal study of biopsies from 120 kidney pancreas patients, and described an increasing presence of arteriolar hyalinosis over time.¹⁹ Such changes have been detected in up to 42% of protocol biopsies,²⁴ identifying a potentially important contributor to chronic transplant dysfunction, which is amenable to therapeutic intervention.

Protocol biopsies may also be a useful tool to detect renal diseases such as BK nephropathy. Activation of BK virus is increasingly common in renal transplant recipients and can lead to a rapid decline in transplant function. Effective and early treatment of BK nephritis by reducing immunosuppressive therapy significantly improves outcome.²⁵

Another potential benefit of protocol biopsies is their implementation in clinical trials designed to prevent CTD. Any such clinical trials are challenging to establish, requiring large numbers of patients to be followed up over a long period. The findings that early diagnosis of TA/IF in early protocol biopsies is an independent predictor of long-term survival, and that such changes are superior to other predictors of outcome, such as acute rejection or serum creatinine, make their use for clinical trials attractive.^26,²⁷

Risks of protocol biopsies

When making a decision about the routine implementation of protocol biopsies after renal transplantation, the risks of the procedure must be taken into account. These have been addressed by Schwarz et al., in a published experience of 1171 protocol biopsies. The complications were gross haematuria (3.1%), perirenal haematoma (3.3%), vasovagal reaction (0.3%) and A-V fistula with spontaneous resolution (9%). There was no associated mortality, and all complications were apparent within the 4-hour post-biopsy surveillance period.²⁸

Protocol biopsies: should they be adopted more widely?

Good evidence supporting the general implementation of protocol biopsies is lacking as large, prospective trials have not been performed. In those studies that have been published, there are many variables, making comparison between studies difficult: the timing of biopsies, immunosuppressive regimens, as well as different histological parameters used to define SCR. In addition, the impact of treatment based on the findings from protocol biopsies can only be fully determined in the long term, and the question as to whether treatment should be instigated on the basis of a single biopsy or sequential biopsies remains unclear.²⁹ Consistent, effective therapeutic strategies for CTD are not available, and this limits the clinical purpose of protocol biopsies in this setting.

An alternative to the histological analysis of protocol biopsies is to adopt molecular technology to seek other strategies for early identification of allografts at risk of rejection.³⁰ For example, high-density array analysis of protocol biopsies demonstrated upregulation of genes associated with inflammation and matrix remodelling, which correlated with histological evidence of TA/IF.³¹ Previous studies implicated growth factors such as transforming growth factor-β (TGF-β), as an important molecule in the development of fibrosis in allografts. Additional non-invasive methods under development seek to examine gene and protein expression profiles in peripheral blood and urine.

The natural history of chronic transplant dysfunction

The chronic changes observed following renal transplantation occur as histological sequelae of a series of pathological insults to the kidney from the time of organ retrieval, which leads to incremental and cumulative damage to nephrons, and ultimate loss of graft function.³² Factors affecting long-term graft survival can be defined by era: peritransplant factors include donor factors, such as age, donor serum creatinine and comorbid conditions such as hypertension; brainstem death with the accompanying catecholamine storm; preservation and implantation injury, e.g. ischaemic damage and reperfusion injury, perhaps leading to delayed graft function. Factors affecting early transplant function include acute rejection and early infection. Over time, recipient factors such as hypertension, hyperlipidaemia and viral infections, e.g. cytomegalovirus, play a more significant role in mediating renal tubular injury. Calcineurin inhibitor toxicity is becoming increasingly recognised as a significant contributor to late injury, giving rise to the paradox that reduction in acute rejection rates has not led to improvements in long-term graft survival.

Until recently it has not been possible to study the progressive changes that culminate in the loss of graft function through tubular atrophy and interstitial fibrosis.

A recent longitudinal study of 961 protocol biopsies performed on 120 renal transplant patients over a period of 10 years has provided information about the natural history of the process.¹⁹

Two distinct phases of injury were evident:

• Early phase of tubulointerstitial injury, with rapidly increasing TA/IF. Early changes of chronic damage were seen at 3 months, and were associated with ischaemic injury, prior severe rejection and SCR.

• Late phase of injury, occurring beyond 1 year, was associated with microvascular and glomerular changes. Progressive high-grade hyalinosis with luminal narrowing, increasing glomerulosclerosis and additional tubulointerstitial damage was accompanied by the use of calcineurin inhibitors. By 10 years, severe evidence of chronic damage was seen in 58.4% of patients, with sclerosis in 37.3% of glomeruli.

Chronic allograft nephropathy: an obsolete term

Our increased understanding of the processes which lead ultimately to chronic allograft failure led to a review of the nomenclature at the eighth Banff conference on allograft pathology held in July 2005.³³ The proposed changes to the diagnostic categories are shown in Box 13.1.

Box 13.1 Banff 07 diagnostic criteria for renal allograft biopsies

Chronic categories are shown in bold. Reproduced from Solez K, Colvin RB, Racusen LC et al. Banff 07 classification of renal allograft pathology: updates and future directions. Am J Transplant 2008; 8(4):753–60. With permission from Blackwell Publishing.

1 Normal

2 Antibody-mediated rejection (AMR)

a Acute AMR

i ATN-like C4d⁺, minimal inflammation

ii Capillary margination and/or thromboses, C4d⁺

iii Arterial-v3, C4d⁺

b Chronic active AMR

Glomerular double contours and/or peritubular capillary basement membrane multilayering and/or interstitial fibrosis/tubular atrophy and/or fibrous intimal thickening in arteries, C4d⁺

3 Borderline changes: suspicious for T-cell-mediated rejection, no intimal arteritis, foci of tubulitis

4 T-cell-mediated rejection

a Acute T-cell-mediated rejection

b Chronic active T-cell-mediated rejection

Chronic allograft arteriopathy

5 Tubular atrophy/interstitial fibrosis (TA/IF), no evidence of specific aetiology

Grade:

I Mild TA/IF (<25% cortical area)

II Moderate TA/IF (26–50%)

III Severe (>50%)

6 Other: changes not considered to be due to rejection

The term chronic allograft nephropathy (CAN) was coined in 1991 as a more generic alternative to the misleading term ‘chronic rejection’, which led to the misconception that all late scarring was due to alloimmune injury.²¹ However, over the years, the term CAN has become used to describe a specific disease entity rather than a descriptive term for non-specific scarring and has led to an acceptance of the inevitability of the process, rather than seeking specific causes. Thus, chronic changes within the kidney should be defined according to aetiology where possible, such as chronic active T-cell-mediated rejection or chronic antibody-mediated rejection. In addition, evidence for CNI toxicity, with arteriolar hyalinosis, should be sought and defined. Other causes of fibrosis and atrophy include recurrent disease, hypertension resulting in glomerulosclerosis, with duplication of internal elastica and arterial fibrointimal thickening, and chronic polyomavirus. Changes detected in specific chronic disease states are shown in Table 13.1. If there is no known aetiology, the term ‘interstitial fibrosis and tubular atrophy, no specific aetiology’ should be adopted.

Table 13.1 Morphology of specific chronic diseases resulting in TA/IF (non-immune mediated)

Aetiology	Morphology
Chronic hypertension	Arterial/fibrointimal thickening with reduplication of elastic, usually with small artery and arteriolar changes
CNI toxicity	Arterial hyalinosisTubular cell injury with vacuolisation
Chronic obstruction	Marker tubular dilatation Large protein casts with extravasation into interstitium and/or lymphatics
Bacterial pyelonephritis	Intra- and peritubular neutrophils, lymphoid follicle formation
Viral infection	Viral inclusions on histology, immunohistochemistry and/or electron microscopy

Aetiological factors

The definition of two phases of injury resulting in chronic changes of tubular atrophy is in keeping with the model of injury in CTD, described in a recent review,³⁴ in which late allograft failure was described as a composite phenotype reflecting the total burden of stressful injury, mediated by five groups of risk factors:³⁵

• donor age;

• brainstem death;

• ischaemia–reperfusion injury;

• immune-mediated injury;

• post-transplant stressors.

Donor age

Donor age has a powerful impact on graft survival. Organs removed from donors at the extremes of age are associated with poorer long-term graft survival compared with young to middle-aged donors.³⁶ This may be related to the reduction in the transplanted nephron mass. Long-standing donor stresses such as hypertension and diabetes also affect long-term outcome.

Other donor risk factors include death from an acute cerebral haemorrhage, female gender (again perhaps due to nephron mass) and Afro-Caribbean ethnic origin.¹

Brainstem death

Donation after brainstem death (rather than live donation) is significantly associated with poor long-term graft survival and an increased incidence of delayed graft function and acute rejection.³⁷ Although the condition of brainstem death is well defined legally and neurologically, the systemic sequelae on the potential donor are very poorly understood. The effects of an irreversible and catastrophic injury to the brainstem include labile blood pressure, alterations in thermal regulation, endocrine and biochemical derangements plus pulmonary changes. Surges in catecholamine release are experienced, with resultant physical and structural changes affecting the vital organs. It may also be the trigger for the systemic release of a range of cytokines, causing a greater exposure of antigenic surface molecules, possibly as a result of endothelial cell retraction. Thereby the transplanted organ may prove more immunogenic to the recipient as a result.