The Color of Dust

In 1967 and ’68, in the hills and valleys of Quang Tri Province, U.S. Marines established positions along the demilitarized zone that separated South Vietnam from the communist North. The Marines did not know the terrain—the jungles, ridges, and ravines of the Annamite Range, where the 60-foot tree canopy and dense undergrowth limited visibility to just a few yards. There, in the far north of South Vietnam, the cold, steady monsoon rains, called the crachin, from the French word for “drizzle,” continued for months. The land and climate were suited to the ancient military strategy of the Vietnamese, used in turning back the Mongols, the Ming, and the French: violent, close-quarters combat. Hundreds of Marines and thousands of their Vietnamese allies died in month after month of intense fighting. In the Situation Room in the White House basement, President Lyndon Johnson followed the Battle at Khe Sanh on a sand-table model. When the Marines finally withdrew in July 1968, the North Vietnamese, bloodied but still formidable, seized control of the area. Nothing had been achieved.

The medical oncologist at the VA called. “Mr. A is back in the hospital,” he said. “They’re calling recurrence, on the brain MRI, in the left temporal lobe, exactly where he was treated with radiation last year.”

“That stinks,” I said. “We can see him.”

I could picture Mr. A in front of me, with his handlebar mustache, its tips twirled into points. He always wore a white T-shirt, a black leather vest, and an army-green “boonie” hat with the sides turned up to display three small enamel pins: one with the “M.I.A.” logo, white letters on a black shield; next to it, the eagle, anchor, and globe insignia of the Marines; and a miniature South Vietnamese flag, with its yellow and red horizontal stripes. He had a welcoming, wide-open grin that he flashed easily and often, and that made him less daunting to talk to, especially in those moments—we’d had many of them—when we discussed matters of life and death. Eight years earlier, he’d been operated on for an early-stage lung cancer. A year after that, he developed an inoperable recurrence, which was treated with radiation combined with chemotherapy. Since then, each of his surveillance CAT scans had looked alike—there was a small chance that he had been cured. After together reviewing the images from the five-year scan, I walked him and his wife out of the exam room.

“We’ll see you next year, with new scans,” I said.

“I’m blessed,” he replied. “Every day.” He nodded to his wife, who had continued walking down the hall. That’s when I first noticed the gold chain he wore, with three small gold trinkets. When he saw what I was looking at, he said, “This one’s a bayonet,” and he showed me the little figurine. “And this one’s a map of the island of Puerto Rico,” he said, turning over a second trinket. The third piece hanging from the necklace was a three-dimensional representation of the eagle, anchor, and globe.

“You have the same pin on your hat,” I said.

“Marine Corps,” he said. “It’s why I got PTSD.” After a pause, he continued, “I was 19 years old, just out of high school. I was bad … baaaad. After my girlfriend broke up with me, I didn’t care about nothing.”

“That’s who they wanted.”

He nodded. “On the plane over there, we were hugging each other, crying, yelling. We knew we were going to kill or be killed. When it was time to land, the airfield was under attack, and the plane was diverted. That’s when we knew this was a real war going on.”

“Unbelievable.”

“Right when the hippies said, ‘It ain’t our war,’ and Muhammad Ali said he wasn’t going. I was only in country a month when a grenade went off in my face. We came upon an NVA camp. The guys were placing satchel charges, to blow it up. I was moving a woodpile, and when I picked up a stack of wood, I heard click.”

“How many guys were you with?”

“Platoon. We got out of there. One guy was giving me a bear hug.” He lowered his voice to a whisper. “I said, ‘Motherfucker, don’t give me a hug, let’s get out of here.’ VC was all around. We could hear them.

“For the next 30 years, liquor was always on me. I was freebasing cocaine, walking around the house with a machete. ‘Get me my rifle,’ I said to her.” He motioned to his wife.

“We have to go,” she said from down the hall.

“What about work?” I said.

“Got high on the job,” he said. “I worked at a chemical solvents plant.”

I shook my head.

“Now it’s 19 years clean,” he said. “Free of alcohol, drugs, and work.” He smiled.

“I’m leaving,” his wife said. She was at the door, ready to exit the clinic.

He nodded toward her. “She gave me an ultimatum. I went in the psychiatric hospital, got therapy for PTSD. She saved my life two or three times. I couldn’t go to the Wall before that. I saw my cousin’s name on the Wall, broke down crying. Even now, if I get sad, I find a place to be alone in the house to cry.” He took a breath, gazed past me. Five years previously, he’d been in the radiation oncology department every day for six weeks, receiving radiation treatment for lung cancer. I had not known that a grenade had gone off in his face, or that he’d walked around the house with a machete. “I’m blessed,” he said.

One morning, six months later, Mr. A woke up with problems finding words. Was it a stroke? His wife took him to the closest hospital—not to the VA, where he’d been treated for cancer—and a brain scan revealed what looked like a brain tumor in the left frontal temporal region, the neuroanatomic area responsible for language. A neurosurgeon removed the tumor, and a week later, the pathology report returned: lung cancer metastasis. After this type of surgery, the chance that the metastasis will return in the very same area in the brain is at least 50 percent. Cancer care guidelines recommend radiation to the surgical region to reduce that risk to about 10 percent.

Just after Mr. A was discharged from the hospital, his wife passed away. He skipped his follow-up appointment, and his attendance became unpredictable—consistent with his years-long approach to medical appointments, many of which he missed without explanation. This was not unusual at the VA, where many of the patients lead complex lives, or live in ways accountable to no one, whether by affected pride or bona fide iconoclasm. For the past five or six years, Mr. A had followed medical recommendations most of the time—enough of the time, apparently: he was alive more than five years after a diagnosis of recurrent advanced lung cancer, which, at the time, fewer than 20 percent of patients survived. Yet by the time he made it to a rescheduled appointment in the radiation oncology clinic, almost six months had elapsed since the operation. Updated scans of his brain and body showed that the tumor had not returned, either in his brain or anywhere else.

When I asked him about what had happened—the emergency rush to the other hospital, the urgent operation—he wasn’t sure of the details. In particular, I wanted to know about the radiation—where and when did he receive it? “You would remember it,” I said. “It’s usually just one day of radiation treatment, with your head locked into an immobilization device—not like with us years ago, to the chest, daily radiation for six weeks.”

“I remember that,” he said. Although he’d presented with word-finding problems, since the operation and the intervening several months, his language fluency had returned to almost normal.

“That’s what I mean.”

He smiled and shook his head. “The brain, I really don’t remember. I wasn’t feeling so well. Then my wife died—she used to take care of all that stuff.”

Eventually, records from the other hospital indicated that Mr. A never underwent radiation after the brain operation. It was an example of a common problem, for patients at the VA or any hospital: when the care becomes fragmented, pieced together from too many physicians, too many hospitals, with no synchronization of the overall plan or goal, important parts can be missed.

As he’d still not received the recommended treatment, he was referred to the university hospital across the street for a specialized radiation procedure called stereotactic radiosurgery, which was not available at the VA. “The data to support radiation in this situation say treat within six months,” said my colleague who saw Mr. A. “He’s now almost eight months out from surgery. It’s also atypical that he developed brain metastases but no metastases anywhere else, almost five years after his chest was treated. I don’t know if he’d still benefit. Plus, his language fluency has returned to just about baseline. Why put him through the risk? Maybe, this far out, he’s proven he won’t recur.”

“Maybe he’s in the 50 percent who don’t recur, even without radiation,” I said.

“Right,” he said. “We could watch it. If it comes back, we could treat it then.”

It came back three months later. In the left frontal temporal lobes, exactly in the region of surgery, a surveillance MRI showed “enhancing nodules,” as the neuroradiologist described them: on the images, white flecks against the gray of the brain, adjacent to the surgical region. “I had thought we might get lucky,” said my colleague from the university hospital. “Still, it’s not so bad. We can treat him now.”

Stereotactic radiosurgery is typically given over just a few hours. Most of that time the patient waits, with his head immobilized to the treatment table, while the treating physicians select the appropriate radiation dose and target and use computer planning to deliver that dose. Mr. A had no neurological sign or symptoms from the tumor—this time, his speech had remained intact; recurrence was detected on an MRI. In the days and weeks after radiosurgery, he had no side effects from the treatment.

“We’ll get another brain MRI in three months,” I told him.

Three months later, the MRI report merely recommended continued surveillance—a promising sign. But at the next three-month scan, the MRI’s intravenous contrast “enhanced” in a way suggestive of cancer recurrence. “Increased vascularity,” apparent as white enhancement on an MRI because of relatively increased contrast uptake, can signify a tumor; it can also be a sign of the inflammatory effects of treatment. Indeed, on this scan’s report, the radiologist wrote, “May represent treatment effect. Cannot rule out recurrence.”

I called my colleague who’d managed the radiosurgery, asked him to look at the MRI with the radiation oncologists and neurosurgeons who’d treated Mr. A. Their impression was that the area of increased vascularization, concerning to the neuroradiologist, likely was not a recurrence. “Maybe get another MRI in two months, instead of three,” my colleague said. “It looks like radiation necrosis.” This problem—complication—means death of the soft tissues. The appearance of radiation necrosis on the MRI, and even the signs and symptoms in the patient, can look very similar to tumor recurrence. Radiation necrosis can even be a life-threatening complication. If the patient is symptomatic, the treatment is a course of steroids, to reduce inflammation, and if that doesn’t work, an operation is required to remove the necrotic tissue causing the symptoms.

Six weeks later, Mr. A woke up one morning and, once again, could not find the words to say anything. His wife gone, his caretaker was now a friend, Tom, who accompanied him to all his appointments and helped him keep those appointments straight. If all patients had medical friends like him, none of them would miss appointments. “It didn’t seem like a stroke to me,” Tom later told me. “But he was out of it. I knew I had to get him to a hospital. And I knew all his care was at the VA, so I brought him here.” Tom could have been a public health expert, I thought, or a health services researcher—he knew the value of continuity of care. When Mr. A arrived at the VA emergency room, a brain MRI was obtained. This time, the neuroradiologist was more certain. “Recurrence,” he wrote, “in the left temporal region, at the site of surgery and radiation.”

This was when the medical oncologist called to alert me to Mr. A’s hospital admission. He’d already been in the hospital a day and a half when I saw him in his room, where he greeted me with a buoyant wave. Steroids had been administered after his brain MRI report had called recurrence. Almost always, metastases are accompanied by surrounding edema—an inflammatory reaction, the brain’s response to a “foreign body.” Steroids work immediately to decrease the swelling, and many patients improve right away. I asked Mr. A to walk back and forth, stand on his toes, and touch his fingertip to his nose. I pushed and pulled on his arms and legs. Nothing was wrong with him—except that he couldn’t speak.

“This is how it’s been,” said Tom, who was sitting in the corner. “He’s gotten much better with the steroids. He cooperates with the doctors, and I can tell he follows what they’re saying.”

Mr. A nodded and smiled. The loss of capacity to speak while maintaining the capacity to understand, and even communicate with gestures, is known as expressive aphasia, or Broca’s aphasia, after the 19th-century French anatomist who first identified it. Paul Broca inferred that language is localized to the brain’s left frontal temporal region—something he arrived at in the course of taking care of Monsieur Leborgne, who came to be called “Tan,” the one word he could articulate; Tan understood language perfectly well and could communicate with gestures, but he could not produce fluent speech. When Tan died in April 1861, Broca found that his patient’s left frontal temporal region was atrophied—the effects, apparently, of chronic vascular injury. In his published report, Broca wrote, “It is … incontestable that this man was intelligent, that he could reflect, and that he had preserved, in a certain measure, his memory for things past.”

“He can say a word here or there,” Tom said. “Maybe it’s better today than it was yesterday.”

Again Mr. A nodded, and smiled.

Then Tom looked around the room—to the ceiling, the windows, the baseboard from one corner to the other. I thought he was about to say how dark the room was, or that he’d lost his wallet. Instead, he said, “I guess it was almost 50 years ago, I was in this hospital for a month, with infectious hepatitis, from malaria. I was yellow.” He stopped his head from wandering the room and returned his gaze to me. “I didn’t know it because all of us were the same color: dust. In the dry season, you couldn’t tell the Black guys from the white guys—everybody was covered with dust. The only way you could tell the Black guys was by their hair. I went from the MASH unit to Fort Bragg to here. In the MASH unit, everybody was trying to get out of there. It was a bad place. Shooting themselves in the foot, or just skinning the leg. My eyes were yellow.”

I looked to Mr. A. “Did you guys know each other?”

He shook his head.

“No, he was two miles away,” Tom said. After a pause, he continued, “He was in a monumental bad place, a memorably bad place, in the history books a bad place.”

“Khe Sanh,” Mr. A said. They were the first words I’d heard him say that day.

“We were a few hills away from each other,” Tom said.

“Thirteen months of dust,” I said.

“Five months of dust, and eight months of mud,” he said. “Tanks would get stuck in it.” He shook his head. “Ugh.”

Mr. A shook his head from side to side.

I took a breath. “This time the MRI says recurrence. Treatment for that is surgery again, or radiation again. Either way, it’s very high risk, including the risk of death from complications from either approach. Let me talk with my colleagues who managed the stereotactic radiosurgery. I’ll get in touch with you next week. Meanwhile, stay on the steroids. Whether it’s recurrent tumor or radiation necrosis, an inflammatory effect of the treatment itself, steroids is the first treatment. Let’s talk next week.”

Tom said, “Sounds good.”

Mr. A nodded his head. On my way out of the room, when I turned from the threshold to look at him, he brought two fingers to his forehead, snapped them forward, and broke into his trademark smile.

A few days later, I talked with my colleague who’d managed the stereotactic radiosurgery. “We looked at it with the neurosurgeons,” he said. “This is radiation necrosis.”

“I don’t see how the neuroradiologist last week sounded so certain, in his report,” I said.

“We see this all the time,” he said. “It may be something that a general neuroradiologist may not have experience with. It’s also a somewhat new phenomenon, that people are living long enough to get this complication. We used to say three to six months for brain metastases. This guy’s been alive two years. That’s in addition to the five years he was alive after inoperable recurrent lung cancer. It’s an unusual story. Let’s keep him on steroids, see what happens.”

I called Mr. A to let him know. The week before, when he’d been in the hospital, we had recommended a second course of radiation to his brain—a highly risky proposal, I’d said, but without it, he could die from recurrent tumor. Now I wanted to tell him he could die from the treatment. Neuroradiologists and neurosurgeons, radiation oncologists and neuro-oncologists, experts in their fields, could not agree on what the images from the MRI meant. Perhaps we, the patient’s physicians, were no more certain than if we were moving pieces around a sand-table model back in LBJ’s Situation Room. Treatment, no treatment, risk of death either way—what was real, and what did any of it mean? We presumed to know what we were doing, but did we even know the terrain?

Aristotle regarded the brain as less important than the heart—the place where spirits gathered, he said, as aisthesin koinen, “common sense.” The Roman physician Galen, connecting brain injuries with changes in cognitive ability, inferred that the brain, not the heart, was the organ for mental activity. He taught that the rete mirabile, the “wonderful net,” in the base of the brain, was where the vital spirit transforms into the animal spirit, which in turn is stored in the brain ventricles—the open spaces under the layers of white and gray matter where the cerebrospinal fluid circulates. “I have dissected more than a hundred heads,” wrote the anatomist Berengario da Carpi in 1521. “I believe that Galen imagined the rete mirabile but never saw it.” His contemporary Leonardo da Vinci created a cast of cerebral ventricles by injecting hot wax into the ventricles of an ox’s brain, then peeling off the brain matter. The anterior, middle, and posterior ventricles were seen as the places for common sense, cogitation, and memory, respectively. More than 300 years later, around the time Broca ascribed language localization to the left side of the brain, a contemporary wrote, “Memory is seated in the posterior part of the eye socket.”

What is language for?

What is the Wall for? What does it mean, and how does it feel, to see a cousin’s name there?

Where is common sense?

Where was common sense at Khe Sanh?

What does it mean to remember a place where it rained for months, where tanks were stuck in the mud, where soldiers shot themselves to get out?

In the Iliad, at the death of Patroclus, “glorious Achilles’s faithful friend,” his fellow soldiers are overcome with grief: “appalled and sick at heart, Antilochus lost for a time his power of speech: his eyes brimmed over, and his manly voice was choked.” What does it mean to be a friend? “A black stormcloud of pain shrouded Achilles. … In the dust he stretched his giant length and tore his hair with both hands.” Why would a man wear the insignia of the U.S. Marines, after his war experience and its aftereffects almost killed him?

When I called Mr. A, Tom answered. He said Mr. A was feeling well, had no pain, was in a good mood. “We thought he couldn’t talk, couldn’t understand us, and he seemed to be getting frustrated. Then a funny thing happened. When he was in the hospital, a few days after we met with you, physical therapy was requested. And the physical therapist, she happened to speak Spanish. When she talked with him, suddenly he became more engaged. I didn’t understand any of it, because I don’t speak Spanish, but up and down the hall, he talked with her. He seemed to have no problems at all. Over the last few days, I’ve been using Google Translate. When I have a question for him, I’ll say it into the phone, and the phone will say it to him in Spanish. Then he’ll reply. I have no idea what he’s saying, but it seems to work.”

I told him that the radiosurgery treatment team had determined that the MRI showed radionecrosis—not recurrent brain metastases but merely a known complication from radiation. The treatment was steroids. “That’s good news,” he said.

I asked Tom to hand the phone to Mr. A. In another life, during a summer away from medical school, I’d traveled in Mexico and South America. Now I summoned my best accent, imagining myself a hapless traveler trying to find the correct bus. “Buenos días,” I began. I asked him whether he found Spanish words easier than English ones. “Yes.” And in every other way he felt fine? I already knew from the scans that there was nothing else wrong with him, at least nothing in the way of lung cancer. He felt well, he said. Then I told him the same things that I’d told Tom: this was a complication, not a recurrence. He had an appointment with neurosurgery in two weeks. He should keep the appointment. I’d given Tom instructions for how he should take the steroids. “Entendiste?” I said.

“Entiendo,” he said.

My mind traveled back to medical school, to the neuroanatomy professor’s first slide, which showed one sentence: “When a song is running in your head, where is it running?” Principles of language organization in the brain follow the broader principles of neuroanatomic localization: for example, the left frontal lobe’s “motor strip” controls right-sided strength and dexterity—handwriting, arm wrestling, swinging a tennis racket, playing the melody line on the piano. That said, neuroanatomic regions subserving one function or another don’t operate so neatly. Writing a postcard isn’t merely moving the hand to write the characters, and returning a tennis ball or playing piano music from memory doesn’t rely only on the motor strip. Neuroanatomic regions typically have multiple functions, which overlap with other neuroanatomic regions. Moreover, neuroanatomic regions are capable of “remodeling.” They are flexible, able to adapt and relearn. A right-handed basketball player can learn to make a lay-up using her left hand.

And so it is with language. Although language is classically placed in the left frontal and temporal lobes, since Broca’s time it has been found that multiple regions contribute to language function—to remembered language, spoken language, understood language. Neurosurgical mapping studies of bilingual speakers have been inconclusive on whether a second language is supported by multiple brain regions or an isolated, localized one. Either way, hypotheses tend to be inconsistent when applied to particular patients.

I didn’t know whether Spanish was Mr. A’s first or second language; my guess was that he used it while growing up in Puerto Rico, or maybe Spanish was the language he had shared with his wife. However he had learned it, the phenomenon of “selective aphasia”—losing the capacity to speak in one language but not another—has been reported in stroke patients and others with neurological problems.

When Tom returned to the phone, he sounded charged, almost exuberant. “That was amazing,” he said. “That’s just how it was when he was talking with the physical therapist.” Because of the uncertainty of his diagnosis, Mr. A had been treated with steroids to reduce inflammation. And with steroids, at least one of his languages had begun to return, slowly. The doctors taking care of Mr. A were convinced that he was getting over a complication from radiation, but the diagnosis was supported only by how well he was feeling. If he again lost the capacity to speak within the next several weeks or months, the team would need to reassess its impression of the MRI.

I reviewed with Tom the time for the upcoming appointment with the neurosurgeon on the radiosurgery team. “You’ll get a call to confirm,” I said. A physician friend has told me that talking with patients about appointment times or scheduling tests with PET scan technicians may not be the best use of my time as a physician. Yes and no. Perhaps E. M. Forster was correct—we physicians need to “only connect,” instead of me trying to explain to the nurse the calls I wanted her to make. Besides, she doesn’t speak Spanish. Fragmented care had gotten us into this situation to begin with. If I could do anything to reduce the risk of missed appointments, lapses in care, I wanted to do it. Before we hung up, Tom said, “Wait—he wants to say one more thing.”

Mr. A returned to the phone. It was something, to realize I’d known him for so many years, known him with and without his wife, through illness and health, and states in between. Ever since he’d told me stories about freebasing cocaine, driving to the Bronx to buy drugs, carrying a rifle around the house, out of his mind, trying to live down the experiences he’d suffered in Vietnam, I’d felt like I was his secret friend. “Thank you, doctor,” he said in English.