HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

This Article Abstract Alert me when this article is cited Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Add to Personal Folders Download to citation manager Author home page(s): K. Eric Sommers Bartley P. Griffith Robert L. Hardesty Robert J. Keenan Permission Requests Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Sommers, K. E. Articles by Keenan, R. J. Search for Related Content PubMed PubMed Citation Articles by Sommers, K. E. Articles by Keenan, R. J.

Ann Thorac Surg 1996;62:784-790
© 1996 The Society of Thoracic Surgeons

---

Original Articles: General Thoracic

Early Lung Allograft Function in Twin Recipients From the Same Donor: Risk Factor Analysis

Division of Cardiothoracic Surgery, University of Pittsburgh, Pittsburgh, Pennsylvania

	Abstract

Top Footnotes Abstract Introduction Material and Methods Results Comment Acknowledgments References

 
Background. Transplantation of lung allografts from the same donor into 2 recipients ("twinning") provides an opportunity to study recipient and donor factors that influence early allograft function.

Methods. Twenty-seven pairs of recipients were identified and evaluated using multivariate logistic regression analysis (p < 0.05). Four measures of early graft function were analyzed: alveolar-arterial gradient in the operating room, first alveolar-arterial gradient in the intensive care unit, alveolar-arterial gradient at 24 hours, and days of mechanical ventilation.

Results. Analysis of the pooled data without regard to pairing showed that alveolar-arterial gradient in the operating room was influenced by donor age, length of donor hospitalization, recipient mean pulmonary artery (PA) pressure at unclamping, and transplantation of a left lung. The alveolar-arterial gradient in the intensive care unit was correlated with donor age, donor cause of death, and mean PA pressure on arrival in that unit. Only mean PA pressure remained significant at 24 hours. Days of mechanical ventilation was determined by mean PA pressure on arrival in the intensive care unit, drop in mean PA pressure during operation, and diagnosis of the recipient. In the paired analysis, receiving a left lung, recipient diagnosis (pulmonary hypertension worse than others), and need of cardiopulmonary bypass were significantly associated with immediate graft dysfunction, although these influences did not persist beyond the immediate postoperative period. Donor arterial oxygen tension and time of ischemia were not significant predictors in any analysis.

Conclusions. Immediate allograft function was associated with donor-related characteristics initially, but these lost importance over the ensuing 24 hours. Recipient PA pressure was an immediate and persisting influence. In the analysis of differences in function between the members of each pair, transplantation of the left lung, recipient diagnosis, and cardiopulmonary bypass were identified, but their influence did not persist beyond the first 6 hours.

	Introduction

Top Footnotes Abstract Introduction Material and Methods Results Comment Acknowledgments References

 
See also page 790.

The transplantation of lungs from the same donor into 2 recipients as single-lung allografts ("twinning") offers an opportunity to examine the influences of allograft factors as well as recipient factors on the early function of the allograft. Although the mechanism of chronic graft dysfunction in lung transplantation is known to be immunologically mediated, the mechanisms underlying early graft dysfunction are not well characterized.

One of the enigmas of clinical lung transplantation is the occurrence of substantial differential graft function in lungs from the same donor implanted in 2 recipients; whereas 1 twin may be extubated early and subsequently enjoy an uncomplicated course, the other may experience severe allograft dysfunction. What factors account for such disparate lung function in organs derived from the same donor? Clearly, function of the newly implanted allograft can be influenced by donor factors (such as age), preservation factors (such as ischemic time), and recipient factors (such as pulmonary artery [PA] pressure).

Prediction of early lung function in single-lung transplantation has proved difficult probably because of the complexity of the interactions between the donor lung and the recipient. We retrospectively reviewed our experience with single-lung allografts placed into separate recipients and used logistic regression to examine the importance of donor and recipient influences.

	Material and Methods

Top Footnotes Abstract Introduction Material and Methods Results Comment Acknowledgments References

 
Data Collection
All pairs who received a lung allograft from the same donor at the University of Pittsburgh were identified retrospectively. Hospital charts and the donor files from the local organ procurement organization were examined, and data on 27 relevant variables were collected. The ideal alveolar gas equation [1] was used to calculate alveolar-arterial (A-a) gradients for oxygen. Care was taken during data abstraction to include measurements during periods of both hemodynamic and ventilatory stability. Five variables were designated as outcome variables: A-a gradient at unclamping (A-aOR), first A-a gradient in the intensive care unit, A-a gradient at 24 hours after unclamping, days of mechanical ventilation, and survival at 12 months. Days in the intensive care unit was initially chosen as an outcome variable but essentially paralleled days of mechanical ventilation and provided no additional information. Because of a declining rate of hospital stay over the period of the study as well as the influence of other factors not related to early graft function, this variable was not considered for outcome analysis.

Surgical Procedure
Details of the surgical implantation procedure have been published previously [2]. In brief, the bronchial anastomosis was performed first with running monofilament suture for the membranous portion of the bronchus and an interrupted telescoping technique for the cartilaginous portion. The arterial anastomosis was performed next with running monofilament, and the venous anastomosis was done last using a similar suture to connect the cuff of the left atrium to the confluence of the native veins at their insertion into the left atrium. Cardiopulmonary bypass (CPB) was used for all patients with pulmonary hypertension and for patients who could not tolerate single-lung ventilation. Intraoperative transesophageal echocardiography was not used early in the series but became routine later to evaluate heart function and venous anastomoses. Immediate portable radionuclide quantitative pulmonary blood flow scanning in the intensive care unit was adopted early in the experience as a routine feature of postoperative care. Any suspicion of an anastomotic complication was vigorously investigated by either immediate reoperation to measure gradients or pulmonary catheterization and arteriography [3].

Donor selection and donor operation have been described previously [4]. In brief, donors were selected on the basis of a clear chest radiography, an arterial oxygen tension higher than 350 mm Hg on an inspired oxygen fraction of 1.0, no history of pulmonary disease or chest operation, appropriate size match and blood type match, and on-site bronchoscopy confirming the absence of aspiration or purulent secretions. Results of donor Gram's staining were not exclusionary unless heavy fungus was present. A history of smoking was not considered a reason for exclusion, and there were no age restrictions in the selection of donors. The donor allografts were procured using a flush with University of Wisconsin solution (60 to 70 mL/kg) through a catheter placed in the main PA and with the lungs inflated with 100% oxygen to an airway pressure no greater than 25 cm H₂O. Prostaglandin E₁ (500 µg) was administered just before aortic clamping. Care was taken to obtain a complete flush of the lungs, which was indicated by blanching of the entire allograft. The lungs were then stored for transportation on ice. Atelectasis was carefully avoided during the period of storage and transportation.

Statistical Methods
Outcome variables were dichotomized at their median values for the unpaired analysis. For the paired analysis, the 2 recipients for each donor were compared with each other for each outcome variable. The member of the pair with the better value for the particular outcome variable was segregated into the "better" group and the other member of the pair, into the "worse" group. This allowed comparable multivariate analysis. Univariate logistic regression analysis was performed, and the results were used in the selection of multivariate models [5]. Analysis of variance was used to compare group means. A p value of 0.05 was considered significant in all analyses.

	Results

Top Footnotes Abstract Introduction Material and Methods Results Comment Acknowledgments References

 
Thirty-five pairs of recipients were identified. Eight pairs were excluded because at least one member of the pair had a criterion for exclusion: known vascular complications (4), intraoperative death not related to donor lung function (2), known fatal brain embolus to the allograft (1), or incomplete records (1).

The study group consisted of 27 pairs of recipients. There were two hospital deaths in this group. One patient died 45 days after transplantation of herpes simplex pneumonitis in the allograft after attempted salvage with extracorporeal membrane oxygenation. The other patient died 32 days after transplantation of graft failure, diffuse alveolar damage, and persistent native lung pneumonia. Another patient with early graft dysfunction was placed on extracorporeal membrane oxygenation for 4 days and was discharged in good condition.

Results of the simple data description for the 27 variables studied are presented in Table 1. Complete information was not available for all variables, and this explains why some do not total 54 (or 27 for donors). The A-a gradient was widest 4 to 6 hours after implantation and also showed the greatest variability at this time point. Although the mean number of days of intubation was 8.25, 26 patients (48%) were intubated for a day or less.

In assessing the influence of donor age, the utility of the multivariate analysis becomes apparent. Figure 1 is a simple scatter diagram of donor age versus A-aOR; there is no evidence of association. After controlling for the other significant predictors, we found a relationship between donor age and A-aOR (here transformed as the model logit) (Fig 2). This comparison demonstrates the shortcomings of simple univariate analysis.

View larger version (10K): [in this window] [in a new window]   Fig 1. . Simple scatter diagram showing no evidence of association between alveolar-arterial gradient in operating room (AaOR) and donor age in unpaired analysis.

View larger version (9K): [in this window] [in a new window]   Fig 2. . Donor age corrected for multivariate influences versus alveolar-arterial gradient in operating room (AaOR LOGIT). Relationship between the two variables is now apparent.

Table 4 presents a subset analysis based on preoperative PA pressure and postoperative allograft perfusion that attempts to more fully elucidate the role of elevated pulmonary resistance in determining allograft function. These results are discussed more fully later.

	Comment

Top Footnotes Abstract Introduction Material and Methods Results Comment Acknowledgments References

Donor age, length of donor hospitalization, and cause of donor death were all associated with early graft function in the first hours after transplantation. Examination of days of donor hospitalization showed that an increase in risk occurs primarily after the first 24 hours. Most donors are mechanically ventilated during this entire period, and this no doubt plays an important role. Colonization of donor airway secretions is known to have an impact on the subsequent development of pneumonia in the recipient (and can be prevented by appropriate antibiotic treatment) [7]. It is possible that the process of mechanical ventilation of itself can be deleterious to lung function.

Donors who die of a closed head injury usually have sustained a blunt force injury resulting from motor vehicle accidents or falls. There are several possible explanations why lungs from these patients may not perform as well in the early postoperative period. Pulmonary contusion, which may not be apparent on chest roentgenogram or visual inspection of the lungs, is one. Fractures of long bones can lead to fat embolus. Closed head injury is commonly associated with intracranial injury and brain swelling with intracranial hypertension, which is known to lead to lung injury in its severest form. These patients may be subject to more vigorous resuscitation, transfusion, and operative intervention for multisystem injury. Patients with closed head injury are also prone to aspiration and development of pneumonia [8]. Clearly, these donor-related variables are markers for cellular and physiologic events that are not fully elucidated.

Although the relationship between donor age and A-aOR is not surprising, the complexity of this association demonstrates that multivariate techniques are necessary to fully assess the impact of a given variable.

Interestingly, ischemic time and best donor arterial oxygen tension were not predictive of outcome in any analysis. Similar findings have been reported by others [9]. These results suggest to us that current techniques of preservation provide sufficient protection during the usual, uncomplicated retrieval and implantation period, as other factors influence early function to a greater degree. Further, we believe newer preservation techniques should address the identification and the modulation of the pathophysiologic state of the donor rather than attempt to focus on lengthening the preservation time of lungs from previously healthy donors as is commonly studied in the laboratory.

The most important recipient factor identified was PA pressure after implantation of the allograft. This factor proved significant from the first time point to at least until the patient was extubated. Recipient PA pressure in patients with a single-lung transplant is determined by the additive resistance of the allograft and the remaining native lung. Thus, elevation of PA pressure can represent a manifestation of allograft injury, residual elevated pulmonary resistance from the remaining native lung, or both (assuming a nonrestrictive anastomosis). Decreased lung function can result from injury as well as hyperperfusion, especially in patients with hypertension. Thus, the interpretation of the pathophysiologic importance of elevated PA pressure and early lung function in single-lung transplant patients is complex.

In an attempt to clarify this, we grouped patients into three categories based on preoperative PA pressure and allograft flow as determined by quantitative radionuclide scanning. Patients with high preoperative PA pressure and hence relatively fixed and elevated resistance in the native lung were compared with patients with relatively low preoperative PA pressure. The low-pressure group was further divided into patients with low flow (<50%) and those with high flow (>50%), the assumption being that patients with high flow would be free from injury (and, on that basis, from elevated resistance). We found that patients who demonstrated less than 50% flow to the allograft (implying elevated resistance in the allograft) had outcomes that were between those of the other two groups. The patients with low preoperative PA pressure and high flow to the allograft had the best outcome measures, although not all the comparisons were significant (see Table 4).

Recipient diagnosis of pulmonary hypertension was also shown to independently predict longer need of mechanical ventilation. This finding is consistent with our earlier investigations [10].

Examination of the results for the paired analysis showed few recipient-related factors that explain marked differential allograft function between pairs. Patients placed on CPB and those with a diagnosis of pulmonary hypertension were shown to be at higher risk for immediate lung dysfunction in the first few hours, but these differences did not persist. This finding is consistent with the well-known detrimental effect of CPB on lung function. In the specific instance of CPB and lung allograft implantation, we [11] have previously shown that patients who require CPB for the implantation of lung allografts have significantly worse early lung function compared with a retrospective control group of patients undergoing transplantation without CPB. As many of the mediators that are known to cause lung dysfunction after CPB have also been identified in association with lung allograft function, it is not surprising that CPB exacerbates early allograft dysfunction. There is information suggesting that patients with pulmonary hypertension are uniquely vulnerable to the damaging effects of bypass [12].

By and large, we were disappointed by our inability to explain differential lung function in our analysis. Of note, the strong influence of postoperative PA pressure in the unpaired analysis was not useful in discriminating the better-performing lungs from the worse-performing lungs in twins. This suggests that donor influences play a relatively stronger role in determining postoperative PA pressure, thereby canceling out recipient influences in the paired analysis. It would have been interesting to have had donor PA pressure data available to attempt to relate them to postoperative pressures. We hypothesize that donor PA pressure may be an important potential predictor of postoperative function, and consideration should be given to routinely obtaining this measurement. Another possible explanation for the absence of predictors in the paired analysis includes exclusion in our data collection of a biologically important variable (such as donor PA pressure).

It is interesting to note the prominent and consistent effect of organ laterality on the first arterial blood gas measurement obtained immediately on reperfusion. There are several potential explanations why the left lung performs worse than the right immediately after implantation. All these operations are done in the lateral decubitus position, and hence, limitations of blood flow may explain why the left lung-having to compete with a larger, dependent right lung for flow-receives less blood flow and less benefit from the new lung. This results in the greater A-a gradient found in left lung recipients. This explanation is supported by the finding that radioisotopic perfusion scans obtained immediately postoperatively showed greater perfusion to right lung allografts than left lung allografts (68% versus 61%; p = 0.15). Another explanation involves the relatively incomplete preservation of the left lung compared with the right, as the left lung is more prone to atelectasis during the retrieval operation.

No predictors that had an impact on survival at 12 months were identified. If the early outcome variables are included as predictors, however, days of mechanical ventilation is highly correlated with 12-month survival (p < 0.008). These results show that although early graft dysfunction can be severe (leading to death or need of extracorporeal membrane oxygenation), its overall influence on long-term survival is not statistically apparent. Infection, acute rejection, and chronic rejection remain the most important threats to survival for lung allograft recipients. The fact that continued need of mechanical ventilation is predictive of survival at 1 year but earlier measures of outcome are not leads us to postulate a "chain of complications" that is put in motion by early lung dysfunction and that sets the stage for pneumonia, acute rejection, colonization with opportunistic organisms, debilitation, and eventually death.

One of the potential confounding problems in this analysis is the possibility of "implantation bias." In other words, were "better" (or what were perceived as better) donors selected for high-risk recipients (those with elevated PA pressure)? If donors who were thought to be "better" were selected for the sickest patients, then there would be a corresponding "leveling" of outcomes, and it might be difficult to demonstrate differences in outcome. To this end, we compared donor age and best donor arterial oxygen tension with preoperative PA pressure and diagnosis. There was no evidence of selection bias in the implantation of lungs from donors with high oxygen tension or young age into patients with a diagnosis of pulmonary hypertension or elevated PA pressure.

We acknowledge several potential limitations to our study, such as the retrospective nature of the data collection and the relatively small size of the sample. The most important implication of these limitations is that our findings may not be generalizable to other institutions or patient samples.

In summary, we used multivariate logistic regression to identify important donor and recipient factors that influence the early function of lung allografts from the same donor placed in different recipients. Donor-related factors prevailed during the first 24 hours, and recipient-related factors-primarily PA pressure-persisted throughout the early period. Survival at 1 year was not associated with any of the early variables studied.

	Acknowledgments

Top Footnotes Abstract Introduction Material and Methods Results Comment Acknowledgments References

 
We acknowledge the helpful suggestions of Marc Schwartz, BS.

	Footnotes

Top Footnotes Abstract Introduction Material and Methods Results Comment Acknowledgments References

 
Presented at the Thirty-second Annual Meeting The Society of Thoracic Surgeons, Orlando, FL, Jan 29–31, 1996.

Address reprint requests to Dr Sommers, University of Pittsburgh, Suite 300 Kaufman Bldg, 3471 Fifth Ave, Pittsburgh, PA 15213-3221.

	References

Top Footnotes Abstract Introduction Material and Methods Results Comment Acknowledgments References

 

1. Nunn JF. Nunn's applied respiratory physiology. 4th ed. Oxford: Butterworth Heinemann, 1993:196.
2. Griffith BP, Magee MJ. Single lung transplantation. Chest Surg Clin North Am 1993;3:75–88.
3. Griffith BP, Magee MJ, Gonzalez IF, et al. Anastomotic pitfalls in lung transplantation. J Thorac Cardiovasc Surg 1994;107:743–54.[Abstract/Free Full Text]
4. Hardesty RL, Aeba R, Armitage JM, Kormos RL, Griffith BP. A clinical trial of University of Wisconsin solution for pulmonary preservation. J Thorac Cardiovasc Surg 1993;105:660–6.
5. Hosmer DW, Lemeshow S. Applied logistic regression. New York: Wiley & Sons, 1989:82–133.
6. Pham SM, Yoshida Y, Aeba R, et al. Interleukin-6, a marker of preservation injury in clinical lung transplantation. J Heart Lung Transplant 1992;11:1017–24.[Medline]
7. Zenati M, Dowling RD, Dummer JS, et al. Influence of the donor lung on development of early infections in lung transplant recipients. J Heart Lung Transplant 1990;9:502–9.
8. Hsieh AH, Bishop MJ, Kubilis PS, Newell DW, Pierson DJ. Pneumonia following closed head injury. Am Rev Respir Dis 1992;146:290–4.[Medline]
9. Glanville AR, Marshman D, Keogh A, et al. Outcome in paired recipients of single transplants from the same donor. J Heart Lung Transplant 1995;14:878–82.[Medline]
10. Bando K, Keenan RJ, Paradis IL, et al. Impact of pulmonary hypertension on outcome after single-lung transplantation. Ann Thorac Surg 1994;58:1336–42.[Abstract]
11. Aeba R, Griffith BP, Kormos RL, et al. Effect of cardiopulmonary bypass on early graft dysfunction in clinical lung transplantation. Ann Thorac Surg 1994;57:715–22.[Abstract]
12. Komai H, Yamamoto F, Tanaka K, et al. Increased lung injury in pulmonary hypertensive patients during open heart operations. Ann Thorac Surg 1993;55:1147–52.[Abstract]

This article has been cited by other articles:

				D. Van Raemdonck, A. Neyrinck, G. M. Verleden, L. Dupont, W. Coosemans, H. Decaluwe, G. Decker, P. De Leyn, P. Nafteux, and T. Lerut Lung Donor Selection and Management Proceedings of the ATS, January 15, 2009; 6(1): 28 - 38. [Abstract] [Full Text] [PDF]

				J. C. Lee and J. D. Christie Primary Graft Dysfunction Proceedings of the ATS, January 15, 2009; 6(1): 39 - 46. [Abstract] [Full Text] [PDF]

				T. Oto, B. J. Levvey, H. Whitford, A. P. Griffiths, T. Kotsimbos, T. J. Williams, and G. I. Snell Feasibility and Utility of a Lung Donor Score: Correlation With Early Post-Transplant Outcomes Ann. Thorac. Surg., January 1, 2007; 83(1): 257 - 263. [Abstract] [Full Text] [PDF]

				I. Al-Githmi, N. Batawil, N. Shigemura, M. Hsin, T. W. lee, G.-W. He, and A. Yim Bronchiolitis obliterans following lung transplantation Eur. J. Cardiothorac. Surg., December 1, 2006; 30(6): 846 - 851. [Abstract] [Full Text] [PDF]

				D.V. Pilcher, C.D. Scheinkestel, G.I. Snell, A. Davey-Quinn, M.J. Bailey, and T.J. Williams High central venous pressure is associated with prolonged mechanical ventilation and increased mortality after lung transplantation J. Thorac. Cardiovasc. Surg., April 1, 2005; 129(4): 912 - 918. [Abstract] [Full Text] [PDF]

				S. Keshavjee, R.D. Davis, M.R. Zamora, M. de Perrot, and G.A. Patterson A randomized, placebo-controlled trial of complement inhibition in ischemia-reperfusion injury after lung transplantation in human beings J. Thorac. Cardiovasc. Surg., February 1, 2005; 129(2): 423 - 428. [Abstract] [Full Text] [PDF]

				M. de Perrot, K. Young, Y. Imai, M. Liu, T. K. Waddell, S. Fischer, L. Zhang, and S. Keshavjee Recipient T Cells Mediate Reperfusion Injury after Lung Transplantation in the Rat J. Immunol., November 15, 2003; 171(10): 4995 - 5002. [Abstract] [Full Text] [PDF]

				M. de Perrot, M. Liu, T. K. Waddell, and S. Keshavjee Ischemia-Reperfusion-induced Lung Injury Am. J. Respir. Crit. Care Med., February 15, 2003; 167(4): 490 - 511. [Abstract] [Full Text] [PDF]

				D. Zaas and S. M. Palmer Respiratory Failure Early After Lung Transplantation: Now That We Know the Extent of the Problem, What Are the Solutions? Chest, January 1, 2003; 123(1): 14 - 16. [Full Text] [PDF]

				W. M. Chatila, S. Furukawa, J. P. Gaughan, and G. J. Criner Respiratory Failure After Lung Transplantation Chest, January 1, 2003; 123(1): 165 - 173. [Abstract] [Full Text] [PDF]

				G. THABUT, I. VINATIER, O. BRUGIERE, G. LESECHE, P. LOIRAT, A. BISSON, J. MARTY, M. FOURNIER, and H. MAL Influence of Preservation Solution on Early Graft Failure in Clinical Lung Transplantation Am. J. Respir. Crit. Care Med., October 1, 2001; 164(7): 1204 - 1208. [Abstract] [Full Text] [PDF]

				D. M. Meyer, L. E. Bennett, R. J. Novick, and J. D. Hosenpud Effect of Donor Age and Ischemic Time on Intermediate Survival and Morbidity After Lung Transplantation Chest, November 1, 2000; 118(5): 1255 - 1262. [Abstract] [Full Text] [PDF]

				J. S. Gammie, D. R. Stukus, S. M. Pham, B. G. Hattler, M. F. McGrath, K. R. McCurry, B. P. Griffith, and R. J. Keenan Effect of ischemic time on survival in clinical lung transplantation Ann. Thorac. Surg., December 1, 1999; 68(6): 2015 - 2019. [Abstract] [Full Text] [PDF]

				F. Venuta, A. Boehler, E. A. Rendina, T. De Giacomo, R. Speich, R. Schmid, G. Funo Coloni, and W. Weder Complications in the native lung after single lung transplantation Eur. J. Cardiothorac. Surg., July 1, 1999; 16(1): 54 - 58. [Abstract] [Full Text] [PDF]

				J. C. Wain, C. D. Wright, D. P. Ryan, S. L. Zorb, D. J. Mathisen, and L. C. Ginns Induction immunosuppression for lung transplantation with OKT3 Ann. Thorac. Surg., January 1, 1999; 67(1): 187 - 193. [Abstract] [Full Text] [PDF]

This Article Abstract Alert me when this article is cited Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Add to Personal Folders Download to citation manager Author home page(s): K. Eric Sommers Bartley P. Griffith Robert L. Hardesty Robert J. Keenan Permission Requests Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Sommers, K. E. Articles by Keenan, R. J. Search for Related Content PubMed PubMed Citation Articles by Sommers, K. E. Articles by Keenan, R. J.