Immunosuppressant infection risk varies by organ and by drug, and a single regimen cannot be judged across all transplant types. In a registry analysis of solid organ recipients, recorded infection was associated with lower long-term survival, but the same drug linked to fewer infections after liver and heart transplant linked to more after kidney transplant. The lesson is that infection risk belongs to a specific drug, dose, and organ combination, not to immunosuppression in general.
The scale of the trade-off is easy to state and hard to manage. Transplant survival has improved sharply, with one-year kidney graft survival moving from roughly 60 to 70 percent to 90 to 95 percent and liver survival from around 20 percent to over 80 percent. That progress depends on suppressing the immune system, and suppression weakens defense against pathogens. Up to 60 percent of recipients develop an infection in the first year, and infection accounts for roughly 12 to 15 percent of deaths after transplantation.
This article sets out the numbers behind that trade-off, the registry method used to examine them, and the drug-by-organ pattern that emerged. It closes with the limits of registry data and what the findings justify at the bedside.
Why infection became the price of transplant survival
The introduction of cyclosporine and tacrolimus changed transplantation. Before calcineurin inhibitors, acute rejection was the dominant early threat. After them, one-year survival rose sharply, and more than 150,000 organ transplants are now performed globally each year. Donor shortage keeps pressure on centers to make each graft last, which means maintaining suppression for years rather than months.
Long-term immunosuppression weakens immune defense. Infection is the most common complication after solid organ transplant and a leading cause of death. The risk tracks the type, dose, and duration of therapy, and it differs across organs because they vary in immunogenicity and clinical tolerability. Kidney and liver grafts are relatively tolerogenic and provoke milder rejection; heart and lung grafts are more immunogenic and demand heavier regimens.
The consequence is a dose-dependent bind. Too little suppression invites rejection and graft loss. Too much invites infection and malignancy. The registry analysis asks whether particular drugs shift that balance in measurable ways once real-world exposure is taken into account.
How the registry analysis was built
The study used a national transplant registry covering multiple organ types and assessed overall survival with Kaplan-Meier analysis, the log-rank test, and Cox regression. Associations between recorded immunosuppressant exposure and a registry-derived binary infection indicator were evaluated with logistic regression, producing odds ratios with confidence intervals.
Element | Detail |
Data source | National solid organ transplant registry across multiple transplant types |
Survival methods | Kaplan-Meier analysis, log-rank test, Cox regression |
Infection methods | Logistic regression on a registry-derived binary infection indicator |
Output measures | Hazard ratios and odds ratios with 95 percent confidence intervals |
Multiplicity | No correction for multiple comparisons, exploratory design |
Significance threshold | Nominal P below 0.05, used as a descriptive reporting threshold |
Software | R version 4.4.1 |
Reporting standard | Strengthening the Reporting of Cohort Studies in Surgery criteria |
The exploratory label matters. P values and confidence intervals were nominal, with no correction for multiple testing, so the findings describe associations in this dataset rather than confirmatory results. A marginal P value is a signal to investigate, not a settled effect.
The survival and infection baseline
The descriptive numbers set the stakes. Infection-related outcomes differ sharply by organ, and the lung recipient carries the heaviest burden.
Metric | Value |
Kidney one-year survival before modern calcineurin inhibitors | roughly 60 to 70 percent |
Kidney one-year survival after calcineurin inhibitors | 90 to 95 percent |
Liver one-year survival before modern calcineurin inhibitors | around 20 percent |
Liver one-year survival after calcineurin inhibitors | over 80 percent |
Recipients with an infection in the first year | up to 60 percent |
Infection as a share of post-transplant deaths | approximately 12 to 15 percent |
Infection as a share of first-year lung transplant deaths | nearly 30 percent |
Infection-related mortality in kidney recipients versus the general population | 7.8 times higher in a regional registry |
Multidrug-resistant gram-negative bacilli in one prospective cohort | rose from 4.8 percent to 38.8 percent |
Registry analysis confirmed the survival link: recipients with a recorded infection had higher mortality and lower overall survival across solid organ types. That result is not surprising by itself. The useful part is the drug-by-organ breakdown, which shows that the immunosuppressant infection risk is not uniform.
The drug-by-organ pattern that emerged
The central finding is that the direction of an association can reverse when the organ changes. Tacrolimus was linked to lower odds of recorded infection after liver and heart transplant but higher odds after kidney transplant. Sirolimus was linked to higher odds after heart, lung, and kidney transplant.
Drug | Organ | Odds ratio for recorded infection | 95 percent CI | P value |
Tacrolimus | Liver | 0.83 | 0.76 to 0.91 | below 0.001 |
Tacrolimus | Heart | 0.91 | 0.87 to 0.95 | below 0.001 |
Tacrolimus | Kidney | 1.58 | not reported in the abstract excerpt | not reported in the abstract excerpt |
Sirolimus | Heart | 1.23 | 1.07 to 1.42 | 0.00454 |
Sirolimus | Lung | 1.80 | 1.28 to 2.57 | below 0.001 |
Sirolimus | Kidney | 1.41 | not reported in the abstract excerpt | not reported in the abstract excerpt |
An odds ratio below 1 points toward lower recorded infection, and one above 1 points toward higher recorded infection. Tacrolimus therefore looks protective in liver and heart recipients and harmful in kidney recipients within this dataset. That reversal defeats any attempt to rank drugs as simply safe or risky.
The registry also captured differences in induction therapy use. Induction was recorded in 30.4 percent of adult liver transplants, compared with 86.3 percent in adult pancreas transplants in the cited comparison. Baseline regimen intensity varies widely by organ, which is one reason a pooled analysis would mislead.
Where mycophenolate mofetil fits in the regimen picture
The major drug classes in the analysis were calcineurin inhibitors such as cyclosporine and tacrolimus, antimetabolites such as azathioprine and mycophenolate mofetil, mTOR inhibitors such as sirolimus and everolimus, glucocorticoids, and cytotoxic antibodies. Mycophenolate mofetil sits in the antimetabolite class, which suppresses lymphocyte proliferation as part of maintenance immunosuppression.
Calcineurin inhibitors work by blocking T cell activation and downregulating cytokines such as IL-2 and IL-4. Tacrolimus prevents rejection better than cyclosporine, with similar survival outcomes. mTOR inhibitors act at a different point in the activation cascade and carry their own infection profile, as the sirolimus results show.
This class context matters for interpreting the numbers. A finding about tacrolimus is a finding about one calcineurin inhibitor, not about every maintenance drug, and a finding about sirolimus does not transfer to the antimetabolite class. Mycophenolate mofetil and other antimetabolites remain part of the background regimen against which these comparisons were made, and the analysis does not isolate their infection effect in the extracted results. Treating the drug classes as interchangeable would be a misreading of the data.
The counterintuitive finding: protective in one organ, harmful in another
The tacrolimus reversal is the most quoted-ready result, and it is also the easiest to misread. The same drug associated with lower recorded infection after liver and heart transplant was associated with higher recorded infection after kidney transplant, with an odds ratio of 1.58 in the kidney group.
Several explanations are plausible. Heart and lung grafts are more immunogenic, so adequate calcineurin inhibition may prevent rejection episodes and the extra immunosuppression used to treat them, indirectly reducing infection exposure. Kidney recipients have different comorbidity profiles, longer expected follow-up, and different attendant care, all of which can change the recorded infection rate. Organ-specific immune microenvironments and differences in clinical tolerability also shape how much drug a patient can receive.
The data cannot choose among these explanations. What they establish is an association that runs in opposite directions by organ. A clinician reading a single pooled estimate for tacrolimus would miss it entirely.
A second counterintuitive point sits in the sirolimus results. mTOR inhibitors are often chosen for their different toxicity profile, yet sirolimus was associated with higher recorded infection across heart, lung, and kidney transplant. The lung estimate carried an odds ratio of 1.80, the largest in the table. That pattern argues against treating an mTOR inhibitor as a default infection-sparing switch.
What the infection spectrum actually includes
The registry indicator records infection as a binary event, which hides the very different organisms involved. Postoperative infections include bacteria, viruses such as cytomegalovirus, Epstein-Barr virus, and BK virus, and fungi such as Aspergillus and Candida. Their incidence and severity track the type, dose, and duration of immunosuppression.
Organism group | Examples named in the source | Relevance to regimen decisions |
Bacteria | Multidrug-resistant gram-negative bacilli | Resistant organisms rose sharply in one prospective cohort |
Viruses | Cytomegalovirus, Epstein-Barr virus, BK virus | Reactivation risk rises with cellular immune suppression |
Fungi | Aspergillus, Candida | Mold disease drives prophylaxis decisions after lung transplant |
Opportunistic | Pneumocystis | Prophylaxis recommended for 6 to 12 months after lung transplant |
The American Society of Transplantation recommends 6 to 12 months of Pneumocystis prophylaxis after lung transplant, reflecting the high infection burden in that population. In a 2017 prospective cohort, multidrug-resistant gram-negative bacilli rose from 4.8 percent to 38.8 percent. Those concrete figures explain why an infection indicator, crude as it is, still carries prognostic weight.
Six practical rules from the registry pattern
These rules are a working interpretation of the analysis. They assume a solid organ recipient on a stable maintenance regimen and do not replace center-specific protocols.
Rule | Statement | Reason in the data |
1 | Read infection risk by organ, not by drug alone | Tacrolimus associations reversed between kidney and other organs |
2 | Treat nominal P values as exploratory | No multiplicity correction was applied |
3 | Weigh regimen intensity alongside the single drug | Induction use varied from 30.4 to 86.3 percent across organs |
4 | Do not assume mTOR inhibitors spare infection | Sirolimus was associated with higher infection in three organs |
5 | Match prophylaxis to the organ's risk profile | Lung recipients carry nearly 30 percent first-year infection mortality |
6 | Keep mycophenolate mofetil in its antimetabolite class context | The analysis did not isolate an antimetabolite-specific infection effect |
Rule 1 is the one most likely to change a conversation. A registry line that reads "tacrolimus lowers infection" is wrong unless the organ is named. Rule 2 keeps the confidence intervals honest, since an exploratory analysis can generate leads but not settle practice.
What this analysis cannot establish
The registry design limits causal inference in several ways. The infection indicator is a binary, registry-derived flag, so it does not distinguish a mild treated infection from a fatal one, and it does not capture timing or organism. Confounding by indication is unavoidable: sicker recipients receive different regimens, and the analysis adjusts only for what the registry records.
The absence of multiplicity correction is another constraint. With several drugs and several organs examined, some nominal associations will arise by chance. The tacrolimus kidney estimate of 1.58 is large enough to deserve attention, but a confirmatory study with a prespecified hypothesis would carry more weight.
Exposure measurement also matters. A recorded prescription is not the same as sustained drug exposure, and trough levels, adherence, and dose changes over follow-up are not captured in a binary exposure field. For drugs such as tacrolimus, where trough level correlates with both rejection and toxicity, that gap is substantial.
Finally, the analysis describes associations with overall survival and a recorded infection flag. It does not show that changing a regimen to avoid the flagged drug would improve outcomes. The right conclusion is that immunosuppressant infection risk is organ specific and deserves to be read that way, not that any particular drug should be abandoned.
Bottom line: immunosuppressant infection risk is organ specific
Immunosuppressant infection risk is a property of a drug, a dose, and a transplanted organ, not of immunosuppression as a whole. Recorded infection was associated with reduced long-term survival across solid organ types in this registry, but the drug signals split by organ: tacrolimus was associated with lower infection after liver and heart transplant and higher infection after kidney transplant, while sirolimus was associated with higher infection after heart, lung, and kidney transplant.
The practical takeaway is to stop asking which drug is safest in the abstract. Ask which drug, at what exposure, in which organ, with what prophylaxis plan. Mycophenolate mofetil and the other antimetabolites remain part of the maintenance background, and the extracted results do not assign them a separate infection signal. The numbers support organ-specific caution and confirmatory study, not a wholesale reordering of maintenance regimens.
FAQ
What is immunosuppressant infection risk?
It is the chance of developing an infection that is linked to the drugs used to prevent transplant rejection. The risk depends on the drug class, the dose, how long therapy continues, and which organ was transplanted. Because these drugs deliberately weaken immune defense, infection becomes one of the main complications of long-term transplant care, alongside rejection itself.
Why do transplant recipients need lifelong immunosuppression?
The immune system recognizes a donated organ as foreign and will attack it without ongoing suppression. Early after transplant, the risk of acute rejection is highest, so regimens are most intensive. Over time the dose is usually reduced, but most recipients stay on some maintenance therapy to protect the graft. That prolonged exposure is what makes infection risk a long-term concern.
Which infections are most common after solid organ transplant?
The spectrum includes bacteria, viruses such as cytomegalovirus, Epstein-Barr virus, and BK virus, and fungi such as Aspergillus and Candida. Which organism matters most depends on the organ, the intensity of suppression, and the time since transplant. Lung recipients face the heaviest burden, with infection accounting for nearly 30 percent of first-year deaths.
Did this study show that one drug is safer than others?
No. The drug associations changed direction by organ. Tacrolimus was linked to lower recorded infection after liver and heart transplant but higher infection after kidney transplant. Sirolimus was linked to higher infection after heart, lung, and kidney transplant. A single ranking of drugs would hide those differences.
What does an odds ratio below 1 mean in this context?
An odds ratio below 1 means the drug was associated with lower odds of recorded infection in that organ group. For example, tacrolimus after liver transplant carried an odds ratio of 0.83. An odds ratio above 1 means higher odds, such as sirolimus after lung transplant at 1.80. These are associations, not proof that the drug caused the change.
Where does mycophenolate mofetil fit in these regimens?
It belongs to the antimetabolite class of maintenance immunosuppressants, alongside azathioprine, and works by limiting lymphocyte proliferation. It was part of the background regimen in this analysis, but the extracted results did not isolate a separate infection signal for it. The findings about tacrolimus and sirolimus should not be assumed to apply to antimetabolites.
How reliable are the study's P values?
The analysis was exploratory and applied no correction for multiple comparisons, so the P values are nominal and used as a descriptive threshold. That means some associations could be due to chance. The results are useful for generating hypotheses and highlighting organ-specific patterns, but they are not confirmatory evidence that would change practice on their own.
Can registry data prove that a drug causes infection?
No. A registry records what happened, not why. The infection flag is binary, so it does not capture severity, timing, or organism, and the data cannot fully adjust for differences between patients that influenced regimen choice. Confounding by indication is a real limitation in this kind of study.
What is Pneumocystis prophylaxis and who needs it?
It is preventive treatment against a fungal organism that can cause severe pneumonia in immunosuppressed patients. The American Society of Transplantation recommends 6 to 12 months of prophylaxis after lung transplant because infection is the leading cause of death in that group. Similar strategies are used in other recipients depending on their regimen and risk.
Should a patient change regimen based on these findings?
Not on this analysis alone. The study describes associations in registry data and did not test whether switching drugs improves outcomes. Regimen decisions depend on rejection history, kidney function, drug levels, comorbidities, and center protocols. The useful message is to discuss infection risk in terms of the specific drug and organ rather than in general terms.
Why do heart and lung grafts need stronger suppression?
These organs are more immunogenic than kidney and liver grafts, meaning they provoke a stronger immune response. That requires heavier immunosuppression to prevent rejection, which in turn raises infection risk. The trade-off is most visible in lung transplant, where infection remains the leading cause of early death.
What is the main takeaway for someone reading this?
The main takeaway is that immunosuppressant infection risk is organ specific. The same drug can look protective in one transplant type and harmful in another. Ask which drug, at what dose, for which organ, and with what prevention plan, rather than asking which immunosuppressant is safest overall.